Power device

Abstract

Power device (100) formed with: - a first main electrode (E1e) on a main surface side of a substrate (7) and - a second main electrode (E1c) is provided on another main surface side of the substrate (7), wherein: - the power device (100) comprises: - a first metal layer (15) formed on the one main surface side of the substrate (7) and electrically connected to the one main surface; and - a second metal layer (12) formed on the first metal layer (15), - the first main electrode (E1e) comprises the first metal layer (15) and the second metal layer (12), - a solder layer (Sd1) is formed on the second metal layer (12), - the power device (100) further comprises an insulating member (X10) surrounded by a first termination (15e) of the first metal layer (15) and a second termination (12e) of the second metal layer (12), the second termination (12e) being opposite the first termination (15e), - the insulating component (X10) is constructed such that the first termination (15e) of the first metal layer (15) is partially connected to the second termination (12e) of the second metal layer (12), - the second metal layer (12) is in direct contact with the first metal layer (15), - the solder layer (Sd1) is in direct contact with the second metal layer (12) and - a recess (V1) is formed in the second termination (12e) of the second metal layer (12) at a region above the insulating component (X10).

Description

Background of the inventionField of the invention

[0001] The present invention relates to a power device provided with electrodes on both surfaces of a substrate. Description of the state of the art

[0002] What is required of a power device is low loss, breakdown voltage withstand capability, an assured safe operation range, reliability, and the like. The safe operation range refers to an operating range for a power device to prevent failure when the power device is in operation. Reliability refers to a property of enduring long-term use. Note that what is further required of the power device is realization of the above plurality of characteristics at the lowest possible cost.

[0003] The criteria representing the reliability required of a power device include thermal cycle resistance, power cycle resistance, and the like. Power devices are used in various environments. Such a power device repeatedly performs, for example, a switching operation for triggering high voltage and large current. In this case, the power device repeatedly encounters a low-temperature state and a high-temperature state.

[0004] It should be noted that a power device is constructed by a plurality of components. Accordingly, when the above-described conditions occur, due to factors including the difference in linear expansion coefficient between a plurality of components constituting the power device, a large force may be exerted on the common portion of the plurality of components. For example, with a power device, metal fatigue may occur at a metal joint portion, resulting in fracture of the metal joint portion. In this case, the function of the power device may cease. In view of the above, a power device is designed to have durability corresponding to the product lifespan.

[0005] A conventional power device often has the following structure to form a current path. In this structure, the lower surface electrode of the semiconductor element included in the power device is connected to one terminal via solder. Further, in this structure, the upper electrode of the semiconductor element is connected to another terminal using a wire made of aluminum or the like.

[0006] With the power device having the above-described structure, in some cases, the heat cycle occurs at the solder connected to the lower surface electrode of the semiconductor element, and the power cycle occurs at the connection area between the upper electrode and the wire. In this case, metal fatigue may occur at the area connected to the wire, thereby limiting or terminating the function of the semiconductor element.

[0007] In recent years, a structure in which solder is connected to the upper surface electrode instead of a wire has been used to increase the amount of current flowing through a power device. Japanese Patent Application Laid-Open No. JP 2015-015395 A discloses a structure in which solder is connected to a surface electrode of a semiconductor element, and the solder is connected to an external terminal (hereinafter also referred to as "Related Structure A").

[0008] In the related structure A, the terminal of an additional electrode is connected to the surface electrode via a protective layer. Note that the protective layer, which is located between the terminal of the additional electrode and the surface electrode, adheres to the terminal of the additional electrode and the surface electrode.

[0009] Accordingly, in the related structure A, when a force is applied to the terminal of the additional electrode connected to the solder, the force tends to be transmitted through the protective layer and the surface electrode to a substrate where a semiconductor element is formed. Accordingly, the related structure A has a problem that when a large force is applied to the terminal of the additional electrode, the substrate is destroyed by the force.

[0010] JP 2006-100530 A relates to a semiconductor device configured to compensate for maximum tensile stress resulting from volume shrinkage with the phase transition of a nickel layer by heat treatment in the case of soldering. The proposed semiconductor device includes a metallic aluminum film formed on a semiconductor and comprising mainly aluminum, and an insulating partition layer dividing the surface of the metallic aluminum film into a plurality of regions. The described semiconductor device further includes a nickel layer formed on the surface of the metallic aluminum film while having boundaries in which a crystal structure is not continued in the upper portion of the insulating partition layer, and comprising mainly nickel.

[0011] JP 2005-183641 A discloses a semiconductor device that can suppress oxidation erosion without significantly changing the electrode boundary structure and that has superior reliability, compact size, and thinness. Two layers of input / output pads are formed on the surface of a substrate, a surface protective film is formed to cover the edge of the input / output pads up to the surface of the substrate, and a metal protrusion electrode is formed on the input / output pads, including the inner periphery of the surface protective film with a barrier metal therebetween. In the semiconductor device having such a structure, a recessed portion such as slits or the like is formed on the inner periphery of the surface protective film to cover the edge of the input / output pads.Thus, a step is created by the slits, and the surface of the barrier metal formed along the step becomes irregularly shaped, and a path for the penetration of water content to a boundary with the barrier metal becomes long. As a result, oxidative erosion of the barrier metal can be suppressed, and at the same time, a contact area between the barrier metal and the protruding metal electrode is increased, thereby preventing the peeling of the protruding metal electrode 6 and thus stabilizing the connection quality.

[0012] JP 2007-208 077 A describes a semiconductor device that can prevent possible breakage of a wiring line due to electromigration. In this known semiconductor device, a first insulating film is provided between a wiring line and a barrier metal. The first insulating film has a first opening where the wiring line faces a bump. The wiring line is electrically connected to the barrier metal through the first opening. A plurality of insulating films are provided in the first opening. A larger number of insulating films are provided on the outer peripheral side of the first opening than on the center side of the opening. Consequently, the electrical resistance on the outer peripheral side of the first opening is greater than the electrical resistance on the center side of the first opening.Thus, the concentration of electromigration on the surface of the bump and a possible break in the wiring due to electromigration can be suppressed.

[0013] WO 2015 / 107796 A1 relates to a semiconductor element having a first main surface and an electrode formed on the first main surface, a protective film formed to cover at least one end of the electrode and having an opening on the surface of the electrode, and a first coating film formed on the electrode in the opening. The protective film includes a side edge portion extending from the side wall of the opening toward the interior of the opening. The side edge portion has an engagement portion for engaging the first coating film. The engagement portion has a surface portion extending in a direction intersecting the first main surface and engages the first coating film because the surface portion contacts the first coating film.

[0014] DE 102014 116 082 A1 discloses a semiconductor device comprising a semiconductor chip with a first main surface and a second main surface. A chip electrode is arranged on the first main surface. The chip electrode has a first metal layer comprising a first metal material selected from the group consisting of W, Cr, Ta, Ti, and metal alloys of W, Cr, Ta, and Ti. The chip electrode further has a second metal layer comprising a second metal material selected from the group consisting of Cu and a Cu alloy, wherein the first metal layer is arranged between the semiconductor chip and the second metal layer. Summary

[0015] An object of the present invention is to provide power devices capable of minimizing the probability of occurrence of a substrate fracture.

[0016] The object underlying the invention is achieved in a power device according to the invention with the features of claim 1, alternatively with the features of claim 3, further alternatively with the features of claim 5 and finally alternatively with the features of claim 6. Advantageous further developments are the subject of the respective dependent claims.

[0017] A power device according to the present invention is a power device provided with a first main electrode on one main surface side of a substrate and a second main electrode on another main surface side of the substrate. The power device includes, among other things, a first metal layer provided on the one main surface side and electrically connected to the one main surface, and a second metal layer formed on the first metal layer. The first main electrode includes the first metal layer and the second metal layer. A solder layer is provided on the second metal layer. The power device further includes an insulating member surrounded by a first termination of the first metal layer and a second termination of the second metal layer, the second termination being opposite to the first termination.The insulating component is constructed such that the first termination of the first metal layer is partially bonded to the second termination of the second metal layer. The second metal layer is in direct contact with the first metal layer. The solder layer is in direct contact with the second metal layer.

[0018] According to one aspect of the present invention, a recess is formed in the second termination of the second metal layer at a region above the insulating component.

[0019] In another aspect of the present invention, the insulating member is formed along an edge portion of the solder layer as viewed in a plan view, the insulating member is constructed by a plurality of insulating layers different in size from each other, and a shape of each of the plurality of insulating layers is ring-like.

[0020] According to a further alternative aspect of the present invention, the insulating member is constructed such that a region in the insulating member closer to an edge of the second metal layer has a higher density.

[0021] In another alternative aspect of the present invention, a Ti layer is formed and the insulating member is disposed on the Ti layer and constructed by an oxide layer or by a nitride layer.

[0022] In the present invention, the first metal layer is provided on one main surface side of the substrate. The insulating member is constructed such that the first terminal of the first metal layer is partially bonded to the second terminal of the second metal layer. That is, adhesion between the first terminal of the first metal layer and the second terminal of the second metal layer is low.

[0023] Accordingly, assuming that a large force is applied to the second termination of the second metal layer where the solder layer is provided, at least a portion of the force is distributed in the direction along the bond plane between the first termination of the first metal layer and the second termination of the second metal layer. As a result, the likelihood of a large force being applied to the substrate is minimized. This minimizes the likelihood of substrate fracture occurring.

[0024] These and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings. Short description of the drawings Fig. 1 is a plan view of a power device according to a first preferred embodiment of the present invention; Fig. 2 is a sectional view of a part of the power device according to the first preferred embodiment of the present invention; Fig. 3 is a sectional view of a part of the power device having the structure according to a first variation of the present invention; Fig. 4 is a plan view of a power device having the structure according to a second variation of the present invention; Fig. 5 is a plan view of a power device having the structure according to a third variation of the present invention; Fig. 6 is a sectional view of a power device having the structure according to the third variation of the present invention; Fig. 7 is a sectional view of a part of a power device having the structure according to a fourth variation of the present invention; Fig. 8 is a plan view of a power device having the structure according to a fifth variation of the present invention; Fig. 9 is a sectional view of a part of a power device having the structure according to the fifth variation of the present invention; Fig. 10 is a plan view of a power device having the structure according to a sixth variation of the present invention. Fig. 11 is a sectional view of a part of the power device having the structure according to the sixth variation of the present invention. Fig. 12 is a sectional view of a part of a power device having the structure according to a seventh variation of the present invention; Fig. 13 is a plan view of a power device having the structure according to an eighth variation of the present invention; Fig. 14 is a sectional view of a part of a power device having the structure according to a ninth variation of the present invention; Fig. 15 is a plan view of a power device as a comparative example; Fig. 16 is a sectional view of a part of the power device as a comparative example; Fig. 17 is a diagram for describing a problem related to the heat cycle; Fig. 18 is a plan view of a power device as another comparative example; and Fig. 19 is a sectional view of a part of the power device as another comparative example. Description of the preferred embodiment

[0025] A description of a preferred embodiment of the present invention will be given below with reference to the drawings. In the drawings referred to below, an identical component is designated by an identical reference numeral. The name and function of components designated by an identical reference numeral are the same. Accordingly, a detailed description of a part of the components designated by an identical reference numeral may be omitted.

[0026] It should be noted that the dimension, material, shape and relative arrangement of components exemplified in the preferred embodiment can be appropriately changed according to the structure of an apparatus to which the present invention is applied, various conditions and the like. <Erste bevorzugte Ausführungsform>

[0027] Fig. 1 is a plan view of a power device 100 according to a first preferred embodiment of the present invention. Fig. 1, an X direction, a Y direction, and a Z direction are perpendicular to each other. In the following drawings, an X direction, a Y direction, and a Z direction are also perpendicular to each other. Hereinafter, a direction including the X direction and a direction opposite to the X direction (-X direction) are also referred to as an "X-axis direction." Furthermore, a direction including the Y direction and a direction opposite to the Y direction (-Y direction) are also referred to as a "Y-axis direction." Furthermore, a direction including the Z direction and a direction opposite to the Z direction (-Z direction) are also referred to as a "Z-axis direction."

[0028] Furthermore, a plane including the X-axis direction and the Y-axis direction is also referred to as an "XY plane." Furthermore, a plane including the X-axis direction and the Z-axis direction is also referred to as an "XZ plane." Furthermore, a plane including the Y-axis direction and the Z-axis direction is also referred to as a "YZ plane."

[0029] Fig. 2 is a sectional view of a part of a power device 100 taken along a line A1-A2 in Fig. 1. With reference to Fig. 1 and Fig. 2, the power device 100 includes an emitter electrode E1e which is a main electrode, a gate electrode E1g, a collector electrode E1c which is a main electrode, a substrate 7, a protective layer 3, a glass coating layer 4, a frame 11, solder layers Sd1, Sd2, and an insulating member X10.

[0030] The emitter electrode E1e is constructed by metal layers 12, 15. That is, the emitter electrode E1e, which is a main electrode, has the metal layer 15 and the metal layer 12. The collector electrode E1c is constructed by metal layers 22, 25.

[0031] As in Fig. As shown in Figure 1, the power device 100 includes a supporting substrate region R3. The supporting substrate region R3 includes a saw line region R2 and a breakdown voltage withstand region R6. The saw line region R2 includes a chip region R1. The chip region R1 corresponds to a part of the supporting substrate region R3.

[0032] The chip region R1 has an emitter electrode region Re and a gate electrode region Rg. The emitter electrode region Re is a region where the emitter electrode E1e is arranged. It should be noted that in Fig. 1 at the emitter electrode region Re the contour of the metal layer 12, which is a part of the emitter electrode E1e, is shown.

[0033] The emitter electrode E1e (the metal layer 12) is connected to the solder layer Sd1. The shape of the solder layer S1d, as seen in a plan view (XY plane), is essentially quadrangular. The gate electrode E1g is arranged in the gate electrode region Rg.

[0034] The substrate 7 is an Si substrate made of silicon. The substrate 7 has a main surface 7a and a main surface 7b. The emitter electrode E1e, which is a main electrode, is provided on the main surface 7a side of the substrate 7. The collector electrode E1c, which is a main electrode, is provided on the main surface 7b side of the substrate 7.

[0035] Each of the metal layers 15, 25 is made of an aluminum alloy (AISi). The metal layer 25 is provided on the main surface 7b of the substrate 7. Each of the metal layers 12, 22 is made of Ni (nickel). Note that the metal layers 15, 25 have a lower hardness than the metal layers 12, 22. The metal layer 22 is formed on the lower surface of the metal layer 25. The metal layer 22 is connected to the frame 11 via the solder layer Sd2.

[0036] The metal layer 15 is provided on the main surface 7a side of the substrate 7. The shape of the metal layer 15, as seen in a plan view (XY plane), is substantially quadrangular. The metal layer 15 is electrically connected to the main surface 7a via a conductive connecting part 6.

[0037] The glass coating layer 4 is provided on the main surface 7a side of the substrate 7. The glass coating layer 4 is provided to cover a part of the main surface 7a of the substrate 7 and a part of the metal layer 15. The glass coating layer 4 functions as a passivation layer. The area where the glass coating layer 4 is present functions as the breakdown voltage withstand region R6.

[0038] As in Fig. As shown in Figure 2, the protective layer 3 is provided on the glass coating layer 4. The protective layer 3 is made of polyimide.

[0039] The metal layer 12 is formed on the metal layer 15. The shape of the metal layer 12, as seen in a plan view (XY plane), is substantially quadrangular. The solder layer Sd1 is provided on the metal layer 12.

[0040] Note that a semiconductor element is formed on the substrate 7, which is disposed under a part of the metal layer 15. The semiconductor element is, for example, an IGBT (Insulated Gate Bipolar Transistor), a MOSFET (Metal Oxide Semiconductor Field Effect Transistor), a diode, or the like. (Characteristic structure)

[0041] Next, a description will be given of the characteristic structure (hereinafter also referred to as "structure Ct1") of the present invention. The structure Ct1 is a structure employing the insulating member X10.

[0042] At the boundary between a termination 12e of the metal layer 12 and a termination 15e of the metal layer 15, the insulating member X10 is independently formed. The insulating member X10 is made of a material that is less adhesive to the metal layer 12 (Ni). Note that the insulating member X10 has a smaller thickness than the metal layer 12. Further, the metal layer 15 and the metal layer 12 are configured to surround the insulating member X10. Specifically, the insulating member X10 is surrounded by the termination 15e of the metal layer 15 and the termination 12e of the metal layer 12. The termination 12e is opposite to the termination 15e. Specifically, the termination 12e is opposite to a part of the termination 15e.

[0043] Furthermore, the insulating component X10 is constructed such that the termination 15e of the metal layer 15 is partially connected to the termination 12e of the metal layer 12. In particular, as shown in Fig. 1, as viewed in a plan view (XY plane), the insulating member X10 is provided along the edge portion of the solder layer Sd1. The insulating member X10 is constructed by a plurality of insulating layers X1 that differ from each other in size.

[0044] As an example, the insulating component X10 is constructed from two insulating layers X1. Fig. 1, broken lines Lc1 represent the contours of the two insulating layers X1. The shape of each of the plurality of insulating layers X1 is ring-like (like a closed loop). Note that the number of insulating layers X1 constituting the insulating member X10 may be one or three or more.

[0045] The manufacturing method of the power device 100 according to the first preferred embodiment is a method that uses a general technique for manufacturing a semiconductor element (deposition, lithography, etching, or the like). Accordingly, a detailed description of the manufacturing method is omitted. A brief description is given below.

[0046] As described above, the thickness of the insulating member X10 is smaller than the thickness of the metal layer 12. Accordingly, as the thickness of Ni is increased by electroless plating or the like, at a time when the thickness of Ni exceeds the thickness of the insulating member X10, Ni begins to grow in the lateral direction. Thus, the insulating member X10 is formed such that the entire insulating member X10 is surrounded by the metal layer 12 and the metal layer 15.

[0047] As described above, according to the present preferred embodiment, the metal layer 15 is provided on the main surface 7a side of the substrate 7. The insulating member X10 is configured such that the terminal 15e of the metal layer 15 is partially bonded to the terminal 12e of the metal layer 12. That is, adhesion between the terminal 15e of the metal layer 15 and the terminal 12e of the metal layer 12 is low.

[0048] Accordingly, provided that a large force is applied to the termination 12e of the metal layer 12 where the solder layer S1d is provided, at least part of the force is distributed in the direction along the connection plane between the termination 15e of the metal layer 15 and the termination 12e of the metal layer 12. As a result, the probability that a large force is applied to the substrate 7 is minimized. Consequently, the probability of fracture of the substrate 7 occurring is minimized. Furthermore, the probability that a large force is applied to the substrate 7 while the temperature of the solder layer Sd1 varies is minimized.

[0049] Here, a description will be given of a comparative example that is the subject of the present preferred embodiment. Hereinafter, a power device as a comparative example will also be referred to as "power device J1." Fig. 15 is a plan view of the power device J1 as a comparative example. Fig. 16 is a sectional view of a part of the power device J1 as a comparative example, taken along a line G1-G2 in Fig. 15.

[0050] The power device J1 differs from the power device 100 in that it does not include the insulating member X10 and the protective layer 3. The rest of the structure of the power device J1 is similar to that of the power device 100, and therefore, a detailed description will not be provided repeatedly.

[0051] In the structure of the power device J1, the solder layer Sd1 is connected to the main electrode (the emitter electrode E1e). This allows increasing the amount of current flowing through the power device J1. However, the power device J1 has a problem of low thermal cycle durability. This problem will be discussed with reference to Fig. 17 described.

[0052] Fig. Figure 17 is a diagram to describe the problem related to the heat cycle. To describe the problem, Fig. 17 shows the state in which a frame 16 is connected to the solder layer Sd1.

[0053] As described above, the hardness of the metal layer 15 is lower than the hardness of the metal layer 12. When the frame 16 expands or contracts due to variations in temperature or the like, a force is exerted on the solder layer Sd1. If the force exerted on the solder layer Sd1 is large, an edge 12ed of the metal layer 12 may promote the occurrence of a crack Wx, as shown in Fig. 17 at the termination 15e of the metal layer 15. In particular, when the force in the vertical direction is large, the crack Wx may also occur on the substrate 7. The occurrence of the crack on the substrate 7 poses a problem that renders the power device J1 unable to drive.

[0054] Further, when the semiconductor element included in the power device J1 assumes an ON state, a force is exerted along both or either of the vertical direction and the horizontal direction on the terminal 12e of the metal layer 12 by the following phenomenon. The phenomenon is, for example, an existence of the difference between the coefficient of linear expansion of the solder layer Sd1 and the coefficient of linear expansion of the frame 16. Further, the phenomenon is, for example, an occurrence of contraction or expansion of the solder layer Sd1 due to variations in a temperature of the solder layer Sd1. When the solder layer Sd1, the metal layer 12, or the like shrinks, a force that separates the terminal 12e of the metal layer 12 from the metal layer 15 occurs at the metal layer 12.

[0055] Further, when an adhesion between the metal layer 12 and the metal layer 15 is large, a force is applied to the metal layer 15, which, as described above, promotes the problem of occurrence of a crack on the metal layer 15.

[0056] To solve this problem, another comparative example can be considered. Hereinafter, a power device as another comparative example will also be referred to as "Power Device J2." Fig. 18 is a plan view of the power device J2 as another comparative example. Fig. 19 is a sectional view of a part of the power device J2 as another comparative example, taken along a line H1-H2 in Fig. 18.

[0057] With reference to Fig. 18 and Fig. 19, the power device J2 differs from the power device J1 in that it further includes the protective layer 3. The rest of the structure of the power device J2 is similar to that of the power device J1, and therefore, a detailed description thereof will not be provided repeatedly.

[0058] With reference to Fig. 19, in the power device J2, the protective layer 3 is provided between the termination 12e of the metal layer 12 and the termination 15e of the metal layer 15. Further, the metal layer 12 is constructed so that the termination 12e of the metal layer 12 converges on the upper surface of the protective layer 3. This reduces the stress occurring at the edge of the metal layer 12 due to thermal deformation of the metal layer 12, the solder layer Sd1, or the like.

[0059] That is, with the power device J2, the thickness of the protective layer 3 must be increased to reduce the stress. Further, the metal layer 12 must be structured so that the termination 12e of the metal layer 12 runs up to the upper surface of the protective layer 3. In the manufacturing method employing plating, in order to achieve the above-described run-up structure, the thickness of the metal layer 12 must be smaller than the thickness of the protective layer 3. In this case, there is a problem that thermal deformation of the metal layer 12 is liable to occur.

[0060] By using sputtering, it is possible to obtain the accumulating structure by reducing the thickness of the metal layer 12 so that it is smaller than the thickness of the protective layer 3. However, sputtering requires the use of a technique such as special masking. Accordingly, the use of sputtering promotes the problem of high manufacturing costs.

[0061] In view of the above, the power device 100 according to the present preferred embodiment has the above-described structure. Accordingly, the power device 100 according to the present preferred embodiment can solve the above-described problems.

[0062] In the present preferred embodiment, as described above, the insulating member X10 is constructed such that the termination 15e of the metal layer 15 is partially bonded to the termination 12e of the metal layer 12. That is, adhesion between the termination 15e of the metal layer 15 and the termination 12e of the metal layer 12 is low.

[0063] Given that a force separating the termination 12e from the termination 15e of the metal layer 15 has occurred at the metal layer 12, the present preferred embodiment correspondingly reduces the average force exerted on the termination 15e of the metal layer 15. Furthermore, the present preferred embodiment improves the thermal cycling durability and the power cycling durability of the power device 100.

[0064] Furthermore, by providing the insulating member X10 at the appropriate position, provided that a crack originating from the metal layer 15 toward the substrate 7 has occurred at the terminal 15e of the metal layer 15, the terminal 12e of the metal layer 12 is separated from the terminal 15e of the metal layer 15 before the crack occurs on the substrate 7. For example, the separation portion (a crack) between the terminal 12e of the metal layer 12 and the metal layer 15 shifts along the horizontal direction. Thus, the structure Ct1 of the present preferred embodiment minimizes the probability of a break (a crack) occurring on the substrate 7.

[0065] Furthermore, the insulating component X10 is constructed by a plurality of ring-shaped insulating layers X1. Accordingly, the speed of the separation portion (a crack) moving along the horizontal direction can be controlled. <Erste Variation>

[0066] Hereinafter, the structure of the present variation is also referred to as "structure Ctm1." In structure Ctm1, the insulating component X10 is constructed of a glass coating layer. Structure Ctm1 is used in structure Ct1 (the first preferred embodiment).

[0067] Fig. 1 is a plan view of the power device 100 having the structure Ctm1 according to the first variation of the present invention. Fig. 3 is a sectional view of a part of the power device 100 having the structure Ctm1, taken along a line A1-A2 in Fig. 1.

[0068] The structure of Ctm1 differs from that in Fig. 2 is precisely that the insulating member X10 is constructed by the glass coating layer 4 instead of the insulating layer X1. The rest of the structure of the structure Ctm1 is similar to that of the structure Ct1, and therefore, a detailed description thereof will not be provided repeatedly. Note that in the structure Ctm1, as viewed in a plan view (XY plane), the insulating member X10 is provided along the edge portion of the solder layer Sd1.

[0069] The shape and structure of the glass coating layer 4 are similar to those of the insulating layer X1. The glass coating layer 4 is also an insulating layer. In the structure Ctm1, the insulating member X10 is constructed by a plurality of glass coating layers 4 (insulating layers) that differ from each other in size.

[0070] Hereinafter, the insulating member X10 constructed by one or more glass coating layers 4 (insulating layers) is also referred to as the "insulating member X10a." As an example, the insulating member X10a is constructed by two glass coating layers 4 (insulating layers). The shape of each of the plurality of glass coating layers 4 (insulating layers) is ring-like (like a closed loop).

[0071] Further, each of the glass coating layers 4 (insulating layers) is an oxide layer, as an example. That is, in the structure Ctm1, the insulating member X10a, which is constituted by a plurality of glass coating layers 4, is constituted by oxide layers.

[0072] Note that the glass coating layers 4 (insulating layers) may be nitride layers, for example. In this case, in the structure Ctm1, the insulating member X10a, which is constituted by the plurality of glass coating layers 4, is constituted by nitride layers.

[0073] The plurality of glass coating layers 4 (insulating layers) constituting the insulating member X10a (X10) are formed as follows. First, a glass film is deposited by deposition or the like. Then, by patterning using lithography, the glass coating layer 4 as a passivation layer and a plurality of glass coating layers 4 constituting the insulating member X10 are simultaneously formed.

[0074] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Furthermore, since the plurality of glass coating layers 4 constituting the insulating member X10a (X10) and the glass coating layer 4 as a passivation layer are formed simultaneously, the manufacturing cost of the power device is reduced. <Zweite Variation>

[0075] Hereinafter, the structure of the present variation is also referred to as "structure Ctm2." In the structure Ctm2, the insulating member X10 is configured by a plurality of insulating parts w1 arranged in a spot-like manner. The insulating parts w1 are made of, for example, a material similar to that constituting the glass coating layer 4. The structure Ctm2 applies to all or part of the structure Ct1 (the first preferred embodiment) and the structure Ctm1 (the first variation).

[0076] Fig. Fig. 4 is a plan view of the power device 100 having the structure Ctm2 according to the second variation of the present invention. Hereinafter, the structure Ct1 (the first preferred embodiment) in which the structure Ctm2 is employed will also be referred to as "structure Ct1m2." Note that Fig. 2 is a sectional view of a part of the power device 100 having the structure Ct1m2, taken along a line B1-B2 in Fig. 4.

[0077] In the structure Ct1m2, each of the plurality of insulating layers X1 constituting the insulating member X10 is constructed by a plurality of insulating parts w1 arranged in a spot-like manner. That is, in the structure Ct1m2, the insulating member X10 is constructed by the plurality of insulating parts w1 arranged in a spot-like manner.

[0078] Furthermore, the structure Ctm1 (the first variation), in which the structure Ctm2 is used, is also referred to as “structure Ctm12”. It should be noted that Fig. 3 is a sectional view of a part of the power device 100 having the structure Ctm12, taken along a line B1-B2 in Fig. 4.

[0079] In the structure Ctm12, each of a plurality of glass coating layers 4 (the insulating layers) constituting the insulating member X10 is constructed by a plurality of insulating parts w1 arranged in a spot-like manner. That is, in the structure Ctm12, the insulating member X10 is constructed by a plurality of insulating parts w1 arranged in a spot-like manner.

[0080] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Further, since the insulating member X10 is constructed by the plurality of insulating parts w1 arranged in a spot-like manner (dot-like), adhesion between the terminal 12e of the metal layer 12 and the terminal 15e of the metal layer 15 can be two-dimensionally averaged. Further, average adhesion between the terminal 12e of the metal layer 12 and the terminal 15e of the metal layer 15 can be equalized. Accordingly, the possibility of occurrence of local stress concentration at the terminal 15e of the metal layer 15 is minimized. <Dritte Variation>

[0081] Hereinafter, the structure of the present variation is also referred to as "structure Ctm3." In the structure Ctm3, the shape of the solder layer Sd1 on its side surface is a concave shape. That is, a concave Ft is formed on the side surface of the solder layer Sd1.

[0082] The structure Ctm3 is used in all or part of the structure Ct1 (the first preferred embodiment), the structure Ctm1 (the first variation) and the structure Ctm2 (the second variation).

[0083] As an example, the structure Ctm12, in which the structure Ctm3 is inserted (hereinafter also referred to as the "structure Ctm123"), is described below. The structure Ctm123 is obtained by inserting the structure Ctm3 into the Fig. 3 shown structure.

[0084] Fig. 5 is a plan view of the power device 100 having the structure Ctm123 according to the third variation of the present invention. Fig. 6 is a sectional view of a part of the power device 100 having the structure Ctm123 according to the third variation of the present invention, taken along a line C1-C2 in Fig. 5. As in Fig. As shown in Figure 6, the shape of the solder layer Sd1 on its side surface is a groove-shaped one. That is, the groove Ft is formed on the side surface of the solder layer Sd1.

[0085] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Furthermore, with the structure Ctm123, the present variation exhibits the effect identical to that exhibited by the second variation.

[0086] Furthermore, since the shape of the solder layer Sd1 on the side surface side is a fillet shape (the shape that spreads downward), a force exerted on the termination 12e of the metal layer 12 is reduced. Accordingly, the possibility of a separation (crack) occurring between the termination 12e of the metal layer 12 and the metal layer 15 is minimized. Consequently, the thermal cycle strength of the power device 100 improves. As a result, the product life of the power device 100 increases. <Vierte Variation>

[0087] Hereinafter, the structure of the present variation is also referred to as "structure Ctm4." In the structure Ctm4, the power device 100 further includes a Ti layer 5. Furthermore, in the structure Ctm4, the insulating member X10 is provided on the Ti layer 5. That is, in the structure Ctm4, the Ti layer 5 exists under the insulating member X10. The Ti layer 5 is made of Ti (titanium).

[0088] The structure Ctm4 is employed in all or part of the structure Ct1 (the first preferred embodiment), the structure Ctm1 (the first variation), the structure Ctm2 (the second variation), and the structure Ctm3 (the third variation).

[0089] As an example, the structure Ctm123, in which the structure Ctm4 is inserted (hereinafter also referred to as "structure Ctm1234"), is described below. The structure Ctm1234 is obtained by inserting the structure Ctm4 into the Fig. 5 and Fig. 6 shown structure.

[0090] Fig. 5 is a top view of the power device 100 having the Ctm1234 structure. Fig. 7 is a sectional view of a part of the power device 100 having the structure Ctm1234 according to the fourth variation, taken along a line C1-C2 in Fig. 5. With reference to Fig. 7, in the Ctm1234 structure, the insulating member X10 is provided on the Ti layer 5. That is, the Ti layer 5 is present under each of the glass coating layers 4 (the insulating layers) constituting the insulating member X10.

[0091] The Ti layer is formed as follows. First, the Ti layer 5 is formed on the metal layer 15 by sputtering. Then, the insulating member X10 is formed on the Ti layer 5 by a general semiconductor device fabrication technique (deposition, lithography, etching, or the like).

[0092] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Furthermore, the structure Ctm1234 exhibits the effect identical to that exhibited by the second and third variations.

[0093] Furthermore, in the Ctm1234 structure, the Ti layer 5 is present under each of the glass coating layers 4 (insulating layers). Accordingly, provided that the pattern in one of the glass coating layers 4 is lost, the state in which adhesion between the termination 12e of the metal layer 12 and the termination 15e of the metal layer 15 is low is maintained.

[0094] Furthermore, due to the presence of the Ti layer 5 on the AlSi metal layer 15, the effect of an AR (anti-reflection) coating on the metal layer 15 is demonstrated. Thus, by designing the metal layer 15 using a cover by lithography, the likelihood of variations in the opening dimension of the cover due to halo effects is minimized.

[0095] Furthermore, due to the presence of the Ti layer 5 under the glass coating layers 4 (the insulating layers), controllability over the dimensions in an exposure production A1 is improved. <Fünfte Variation>

[0096] The structure of the present variation is also referred to below as "structure Ctm5." In the structure Ctm5, a recess V1 is formed in the termination 12e of the metal layer 12 in a region above the insulating component X10.

[0097] The structure Ctm5 is employed in all or part of the structure Ct1 (the first preferred embodiment), the structure Ctm1 (the first variation), the structure Ctm2 (the second variation), the structure Ctm3 (the third variation), and the structure Ctm4 (the fourth variation).

[0098] As an example, the structure Ctm1, Ctm3, in which the structure Ctm5 is used (hereinafter also referred to as “structure Ctm135”), is described below.

[0099] Fig. 8 is a plan view of the power device 100 having the structure Ctm135 according to the fifth variation of the present invention. Fig. 9 is a sectional view of a part of the power device 100 having the structure Ctm135 according to the fifth variation of the present invention, taken along a line D1-D2 in Fig. 8. The structure Ctm135 is obtained by inserting the structure Ctm5 into the Fig. 6 shown structure. As shown in Fig. 9, in the termination 12e of the metal layer 12, the recess V1 is formed at a region above the insulating component X10.

[0100] The shape of the recess V1, as seen in a plan view (XY plane), is ring-like (like a closed loop). Specifically, the recess V1 is formed at the end 12e of the metal layer 12, so that the solder layer Sd1 is provided on the inner side of the ring-like (like a closed loop) recess V1.

[0101] It should be noted that Fig. 8 and Fig. 9 show the state in which the insulating member X10 is constructed by a glass coating layer 4. Fig. 8 shows the arrangement position of the insulating member X10 (the glass coating layer 4) by a broken line Lc2.

[0102] Further, as an example, the glass coating layer 4 (the insulating layer) is constituted by an oxide layer. That is, the insulating member X10a constituted by the glass coating layer 4 is constituted by an oxide layer.

[0103] Note that, as an example, the glass coating layer 4 (the insulating layer) may be constituted by a nitride layer. In this case, the insulating member X10a constituted by the glass coating layer 4 is constituted by a nitride layer.

[0104] In the Ctm135 structure, as seen from a plan view (XY plane), the insulating member X10 is provided along the edge portion of the solder layer Sd1. The shape of the glass coating layer 4 (the insulating layer) is ring-like. Note that the width of the ring-like glass coating layer 4 is set to an appropriate size so that the recess V1 becomes like a closed loop (ring-like).

[0105] It should be noted that similar to Fig. 3, the insulating member X10 may be constructed by a plurality of glass coating layers 4. That is, the insulating member X10 may be constructed by a plurality of glass coating layers 4 (the insulating layers) that differ in size from each other. In this case, the shape of each of the plurality of glass coating layers 4 (the insulating layers) is ring-like.

[0106] Furthermore, in the Ctm135 structure, the shape of the solder layer Sd1 on its side surface is a fillet shape. That is, the fillet Ft is formed on the side surface of the solder layer Sd1.

[0107] Note that since the method for manufacturing the structure Ctm5 is similar to that in the first and second variations, a description thereof is omitted. However, when the width of the ring-like glass coating layer 4 is extremely narrow, the recess V1, which is large in size, is not formed at the termination 12e of the metal layer 12. Further, when the width of the glass coating layer 4 is extremely large, the pattern is separated. Accordingly, the width of the glass coating layer 4 is set to an appropriate size, taking the thickness of the metal layer 12 into consideration.

[0108] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Furthermore, the structure Ctm135 exhibits the effect identical to that exhibited by the third variation.

[0109] Further, in the structure Ctm5, since the recess V1 is provided at the termination 12e of the metal layer 12, when a semiconductor device (a chip) is connected to the solder layer Sd1, the flowing solder is prevented from reaching the edge (the edge 12ed) of the metal layer 12.

[0110] This reduces a force exerted by the edge 12ed of the metal layer 12 on the metal layer 15. Accordingly, the probability of a break (crack) occurring on the substrate 7 is minimized.

[0111] Further, by changing the number and width of the recess, the solder is effectively prevented from reaching the edge (edge ​​12ed) of the metal layer 12. <Sechste Variation>

[0112] Hereinafter, the structure of the present variation is also referred to as "structure Ctm6." In structure Ctm6, the thickness of the insulating member X10 is smaller than the thickness of the metal layer 12. Further, the thickness of the termination 12e of the metal layer 12 gradually decreases toward the edge (edge ​​12ed) of the termination 12e. Furthermore, in structure Ctm6, in the insulating member X10, the insulating member X10 is structured such that a portion in the insulating member X10 closer to the edge (edge ​​12ed) of the metal layer 12 has a higher density.

[0113] The structure Ctm6 is employed in all or part of the structure Ct1 (the first preferred embodiment), the structure Ctm1 (the first variation), the structure Ctm2 (the second variation), the structure Ctm3 (the third variation), the structure Ctm4 (the fourth variation), and the structure Ctm5 (the fifth variation).

[0114] As an example, the structure Ctm1, Ctm3, Ctm5 is described below, in which the structure Ctm6 is inserted (hereinafter also referred to as "structure Ctm1356"). The structure Ctm1356 is obtained by inserting the structure Ctm6 into the Fig. 9 shown structure.

[0115] Fig. 10 is a plan view of the power device 100 having the Ctm1356 structure according to the sixth variation of the present invention. Fig. 11 is a sectional view of a part of the power device 100 having the structure Ctm1356 according to the sixth variation, taken along a line E1-E2 in Fig. 10. As in Fig. 11, in the structure Ctm1356, the recess V1 is formed in the termination 12e of the metal layer 12 at a region above the insulating member X10.

[0116] Furthermore, as in Fig. 10 and Fig. As shown in Figure 11, as viewed in plan view (XY plane), the insulating member X10 is provided along the edge portion of the solder layer Sd1. The insulating member X10 is constructed by a plurality of glass coating layers 4 (insulating layers) that differ in size from each other.

[0117] As an example, the insulating component X10 is constructed by three glass coating layers 4 (the insulating layers). Fig. 10 shows the contours of the three glass coating layers 4 (insulating layers) by broken lines Lc3. The shape of each of the plurality of glass coating layers 4 (insulating layers) is ring-like.

[0118] Further, the width of the plurality of glass coating layers 4 (insulating layers) is set so that the thickness of the termination 12e of the metal layer 12 gradually decreases toward the edge 12ed. Specifically, the closer the glass coating layers 4 are to the edge 12ed, the larger the width of the plurality of glass coating layers 4 (insulating layers). That is, the insulating member X10 is configured so that, in the insulating member X10, a portion closer to the edge (edge ​​12ed) of the metal layer 12 has a higher density. Thus, the thickness of the termination 12e of the metal layer 12 gradually decreases toward the edge (edge ​​12ed) of the termination 12e.

[0119] Furthermore, in the Ctm1356 structure, the shape of the solder layer Sd1 on its side surface is a fillet shape. That is, the fillet Ft is formed on the side surface of the solder layer Sd1.

[0120] The manufacturing method of the structure Ctm6 is realized by shaping the glass coating layers 4, and therefore, the description of the manufacturing method is omitted. Basically, the structure Ctm6 is manufactured such that the plurality of glass coating layers 4 have a larger width as the glass coating layers 4 are closer to the edge 12ed.

[0121] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Further, the thickness of the termination 12e of the metal layer 12 gradually becomes smaller toward the edge (the edge 12ed) of the termination 12e. Accordingly, the edge 12ed of the metal layer 12 reduces a force applied to the metal layer 15. Accordingly, the probability of a break (crack) occurring on the substrate 7 is minimized.

[0122] Furthermore, the Ctm1356 structure exhibits the effect similar to that shown by the third and fifth variations. Accordingly, the thermal cycle durability of the power device 100 improves. As a result, the product life of the power device 100 increases. <Siebte Variation>

[0123] Hereinafter, the structure of the present variation is also referred to as "structure Ctm7." The structure Ctm7 is obtained by inserting the structure Ctm4 into the structure Ctm6. In the structure Ctm7, the insulating component X10 is provided on the Ti layer 5. That is, in the structure Ctm4, the Ti layer 5 is present under the insulating component X10.

[0124] The structure Ctm7 is employed in all or part of the structure Ct1 (the first preferred embodiment), the structure Ctm1 (the first variation), the structure Ctm2 (the second variation), the structure Ctm3 (the third variation), and the structure Ctm5 (the fifth variation).

[0125] As an example, the structure Ctm1, Ctm3, Ctm5, Ctm6, in which the structure Ctm7 is inserted (hereinafter also referred to as "structure Ctm13567"), is described below. The structure Ctm13567 is created by inserting the structure Ctm7 (structure Ctm4, Ctm6) into the Fig. 11 shown structure.

[0126] Fig. 10 is a top view of the power device 100 having the structure Ctm13567. Fig. 12 is a sectional view of a portion of the power device 100 having the structure Ctm13567, taken along a line 12 in Fig. 10 shown line E1-E2. As in Fig. As shown in Fig. 12, the insulating member X10 is provided on the Ti layer 5. That is, the Ti layer 5 is present under each of the glass coating layers 4 (the insulating layers) constituting the insulating member X10.

[0127] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Furthermore, structure Ctm13567 exhibits the effect exhibited by the fourth variation (structure Ctm4) and the effect exhibited by the sixth variation (Ctm6). <Achte Variation>

[0128] The structure of the present variation is also referred to as "structure Ctm8" below. The structure Ctm8 is obtained by inserting the structure Ctm2 into the structure Ctm6.

[0129] The structure Ctm8 is employed in all or part of the structure Ct1 (the first preferred embodiment), the structure Ctm1 (the first variation), the structure Ctm3 (the third variation), the structure Ctm4 (the fourth variation), and the structure Ctm5 (the fifth variation).

[0130] As an example, the structure Ctm1, Ctm3, Ctm5, in which the structure Ctm8 is inserted (hereinafter also referred to as "structure Ctm1358"), is described below. The structure Ctm1358 is created by inserting the structure Ctm8 (structure Ctm2) into the Fig. 10 and Fig. 11 shown structure.

[0131] Fig. 13 is a plan view of the power device 100 having the Ctm1358 structure according to the eighth variation of the present invention. Note that Fig. 11 is a sectional view of a portion of the power device 100 having the structure Ctm1358, taken along a line F1-F2 in Fig. 13.

[0132] As in Fig. As shown in Figure 13, in the structure Ctm1358, the insulating member X10 is constructed by a plurality of insulating parts w1 arranged in a spot-like manner. Specifically, in the structure Ctm13568, each of the plurality of glass coating layers 4 (insulating layers) constituting the insulating member X10 is constructed by a plurality of insulating parts w1 arranged in a spot-like manner.

[0133] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Furthermore, the structure Ctm1358 exhibits the effect exhibited by the second variation. <Neunte Variation>

[0134] Hereinafter, the structure of the present variation is also referred to as "structure Ctm9." In the structure Ctm9, the metal layer 15 is composed of a plurality of layers (components) that differ from one another in hardness. Hereinafter, the metal layer 15 composed of the plurality of layers (components) that differ from one another in hardness is also referred to as the "metal layer 15A."

[0135] The Ctm9 structure is inserted into all or part of the Ct1 structure, the Ctm1 structure, the Ctm2 structure, the Ctm3 structure, the Ctm4 structure, the Ctm5 structure, the Ctm6 structure, the Ctm7 structure and the Ctm8 structure.

[0136] As an example, the structure Ctm1, in which the structure Ctm9 is inserted (hereinafter also referred to as the “structure Ctm19”), is described below. Fig. 1 is a plan view of the power device 100 having the structure Ctm19 according to the ninth variation. Fig. 14 is a sectional view of a part of the power device 100 having the structure Ctm19, taken along a line A1-A2 in Fig. 1.

[0137] With reference to Fig. 14, the metal layer 15A comprises a layer 15b, a layer 15a, and a layer 15c. Each of the layers 15a, 15b, 15c is made of, for example, an aluminum alloy (AISi). The hardness of layer 15b is higher than the hardness of the layers 15a, 15c. That is, the hardness of the layers 15a, 15c is lower than the hardness of the layer 15b.

[0138] Layer 15b is provided on layer 15c. Layer 15a is provided on layer 15b. That is, layer 15b is embedded between layers 15a and 15c. Layer 15a is in contact with metal layer 12. That is, layer 15b is not in contact with metal layer 12.

[0139] The process for fabricating the metal layer 15A is carried out as follows. The layers 15a, 15b, 15c are formed such that the grain size of the layers 15a, 15c is larger than the grain size of the layer 15b.

[0140] Specifically, layers 15a, 15b, and 15c are formed by sputtering. First, layer 15c is formed at an increased sputtering power. Next, layer 15b is formed at a reduced sputtering power. Next, layer 15a is formed by a method identical to that used to form layer 15c.

[0141] As described above, the present variation exhibits the effect identical to that exhibited by the first preferred embodiment. Further, the structure Ctm19 exhibits the effect exhibited by the first variation. Furthermore, in the structure Ctm9 included in the structure Ctm19, the layer 15a is in contact with the metal layer 12. That is, the layer 15b, which has a higher hardness than the layer 15a, is not in contact with the metal layer 12. Accordingly, given that a crack has occurred at the layer 15a of the metal layer 15A, the presence of the layer 15b minimizes the probability of the crack reaching the substrate 7. Accordingly, this further minimizes the probability of a fracture (crack) occurring at the substrate 7.

[0142] Furthermore, the present variation can reduce adhesion between the termination 12e of the metal layer 12 and the metal layer 15A. Accordingly, occurrence of a separation region (a crack) between the termination 12e and the metal layer 15A is enabled, thereby minimizing the possibility of occurrence of a break (a crack) on the substrate 7.

[0143] The present invention is applicable to a power device regardless of the breakdown voltage class of the power device, the type of the substrate (an FZ substrate, an epitaxial substrate), and the like.

[0144] It should be noted that the material for reducing the adhesion between the metal layer 12 and the metal layer 15 is not limited to an oxide layer, a nitride layer, and Ti, and may be another material.

[0145] Note that although the metal layers 22, 25 are provided on the main surface 7b side of the substrate 7, a non-oxidizable material (Au) may be formed on the surface (the bottom surface) of the metal layer 22. This also contributes to improving stability of a power device structure.

Claims

[1] Power device (100) formed with: - a first main electrode (E1e) on a main surface side of a substrate (7) and - a second main electrode (E1c) is provided on another main surface side of the substrate (7), wherein: - the power device (100) comprises: - a first metal layer (15) formed on the one main surface side of the substrate (7) and electrically connected to the one main surface; and - a second metal layer (12) formed on the first metal layer (15), - the first main electrode (E1e) comprises the first metal layer (15) and the second metal layer (12), - a solder layer (Sd1) is formed on the second metal layer (12), - the power device (100) further comprises an insulating component (X10) surrounded by a first termination (15e) of the first metal layer (15) and a second termination (12e) of the second metal layer (12), the second termination (12e) being opposite the first termination (15e), - the insulating component (X10) is constructed such that the first termination (15e) of the first metal layer (15) is partially connected to the second termination (12e) of the second metal layer (12), - the second metal layer (12) is in direct contact with the first metal layer (15), - the solder layer (Sd1) is in direct contact with the second metal layer (12) and - a recess (V1) is formed in the second termination (12e) of the second metal layer (12) at a region above the insulating component (X10). [2] Power device (100) according to claim 1, wherein: - the insulating component (X10) has a smaller thickness than the second metal layer (12) and - a thickness of the second termination (12e) of the second metal layer (12) gradually decreases towards an edge (12ed) of the second termination (12e). [3] Power device (100) formed with: - a first main electrode (E1e) on a main surface side of a substrate (7) and - a second main electrode (E1c) is provided on another main surface side of the substrate (7), wherein: - the power device (100) comprises: - a first metal layer (15) formed on the one main surface side of the substrate (7) and electrically connected to the one main surface; and - a second metal layer (12) formed on the first metal layer (15), - the first main electrode (E1e) comprises the first metal layer (15) and the second metal layer (12), - a solder layer (Sd1) is formed on the second metal layer (12), - the power device (100) further comprises an insulating component (X10) surrounded by a first termination (15e) of the first metal layer (15) and a second termination (12e) of the second metal layer (12), the second termination (12e) being opposite the first termination (15e), - the insulating component (X10) is constructed such that the first termination (15e) of the first metal layer (15) is partially connected to the second termination (12e) of the second metal layer (12), - the second metal layer (12) is in direct contact with the first metal layer (15), - the solder layer (Sd1) is in direct contact with the second metal layer (12), - as seen in a plan view, the insulating component (X10) is provided along an edge region of the solder layer (Sd1), - the insulating component (X10) is constructed by a plurality of insulating layers (X1, 4) which differ from one another in size, and - a shape of each of the plurality of insulating layers (X1, 4) is ring-like. [4] The power device (100) according to claim 4, wherein the insulating member (X10) is constructed by a plurality of insulating parts (w1) arranged in a patch-like manner. [5] Power device (100) formed with: - a first main electrode (E1e) on a main surface side of a substrate (7) and - a second main electrode (E1c) is provided on another main surface side of the substrate (7), wherein: - the power device (100) comprises: - a first metal layer (15) formed on the one main surface side of the substrate (7) and electrically connected to the one main surface; and - a second metal layer (12) formed on the first metal layer (15), - the first main electrode (E1e) comprises the first metal layer (15) and the second metal layer (12), - a solder layer (Sd1) is formed on the second metal layer (12), - the power device (100) further comprises an insulating component (X10) surrounded by a first termination (15e) of the first metal layer (15) and a second termination (12e) of the second metal layer (12), the second termination (12e) being opposite the first termination (15e), - the insulating component (X10) is constructed such that the first termination (15e) of the first metal layer (15) is partially connected to the second termination (12e) of the second metal layer (12), - the second metal layer (12) is in direct contact with the first metal layer (15), - the solder layer (Sd1) is in direct contact with the second metal layer (12) and - the insulating component (X10) is constructed such that a region in the insulating component (X10) which is closer to the edge (12ed) of the second metal layer (12) has a higher density. [6] Power device (100) formed with: - a first main electrode (E1e) on a main surface side of a substrate (7) and - a second main electrode (E1c) is provided on another main surface side of the substrate (7), wherein: - the power device (100) comprises: - a first metal layer (15) formed on the one main surface side of the substrate (7) and electrically connected to the one main surface; and - a second metal layer (12) formed on the first metal layer (15), - the first main electrode (E1e) comprises the first metal layer (15) and the second metal layer (12), - a solder layer (Sd1) is formed on the second metal layer (12), - the power device (100) further comprises an insulating component (X10) surrounded by a first termination (15e) of the first metal layer (15) and a second termination (12e) of the second metal layer (12), the second termination (12e) being opposite the first termination (15e), - the insulating component (X10) is constructed such that the first termination (15e) of the first metal layer (15) is partially connected to the second termination (12e) of the second metal layer (12), - the second metal layer (12) is in direct contact with the first metal layer (15), - the solder layer (Sd1) is in direct contact with the second metal layer (12), - a Ti layer (5) is formed and - the insulating component (X10) is formed on the Ti layer (5) and is constructed by an oxide layer (4) or by a nitride layer (4). [7] The power device (100) according to any one of the preceding claims, wherein a shape of the solder layer (Sd1) on its surface side is a fillet shape. [8] Power device (100) according to one of the preceding claims, wherein: - the first metal layer (15A) comprises a first layer (15b) and a second layer (15a) formed on the first layer (15b), - the second layer (15a) has a lower hardness than the first layer (15b), and - the second layer (15a) is in contact with the second metal layer (12).

Images