Semiconductor device

Abstract

This semiconductor device (10) comprises a first switching element (30A), a second switching element (30B), and a resin layer (50) that seals each switching element (30A, 30B). The semiconductor device (10) includes: a power supply electrode (71) which is formed on the surface of the resin layer (50) and of which at least a part overlaps the first switching element (30A) when viewed from a z direction; and a power supply via conductor (61) that electrically connects the first switching element (30A) and the power supply electrode (71) through the resin layer (50) in the z direction.

Description

Technical Field

[0001] The present disclosure relates to a semiconductor device.

Background Art

[0002] Conventionally, a semiconductor device in which a plurality of semiconductor elements are encapsulated with a sealing resin to form a package is known. For example, the semiconductor device of Patent Document 1 includes, as semiconductor elements, a plurality of switching elements, a driver for driving the plurality of switching elements, a metal lead frame terminal for supporting each switching element and the driver, and a sealing resin for encapsulating each switching element and the driver. The lead frame terminal constitutes an exterior electrode exposed to the outside of the semiconductor device.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in the semiconductor device of Patent Document 1, a plurality of switching elements and a lead frame terminal are electrically connected using wires formed by wire bonding, and each switching element and the driver are electrically connected. In this case, since the wires are formed in a curved convex shape, it is difficult to shorten the length of the wires. For this reason, it is difficult to shorten the conductive path in the semiconductor device, and there is room for improvement in reducing the inductance caused by the length of this conductive path.

[0005] An object of the present disclosure is to provide a semiconductor device capable of reducing inductance.

Means for Solving the Problems

[0006] A semiconductor device for solving the above problems includes a switching element, a specific element, and a resin layer that encapsulates the switching element and the specific element, and is a semiconductor device with the thickness direction of the resin layer as the height direction. An external electrode formed on the surface of the resin layer, at least a part of which overlaps the switching element when viewed from the height direction, and an element driving via conductor that penetrates the resin layer in the height direction to electrically connect the switching element and the external electrode.

[0007] According to this configuration, since the switching element and the external electrode are connected by the element driving via conductor that penetrates the resin layer in the height direction, the length of the conductive path between the switching element and the external electrode can be shortened compared to the configuration in which the switching element and the external electrode are connected by a wire formed by wire bonding. Therefore, the inductance caused by the length of this conductive path can be reduced.

[0008] A semiconductor device for solving the above problems includes a switching element, a driver for driving the switching element, and a resin layer that encapsulates both the switching element and the driver, and is a semiconductor device with the thickness direction of the resin layer as the height direction. The switching element and the driver are arranged at intervals in a direction orthogonal to the height direction, and an element control via conductor for connecting the switching element and the driver is embedded in the resin layer.

[0009] According to this configuration, since the switching element and the driver are connected by the element control via conductor embedded in the resin layer, the length of the conductive path between the switching element and the driver can be shortened compared to the configuration in which the switching element and the driver are connected by a wire formed by wire bonding. Therefore, the inductance caused by the length of this conductive path can be reduced.

Advantages of the Invention

[0010] According to the above semiconductor device, the inductance can be reduced.

Brief Description of the Drawings

[0011]

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Fig. 9

Fig. 10

Fig. 11

Fig. 12

Fig. 13

Fig. 14

Fig. 15

Fig. 16

Fig. 17

Fig. 18

Fig. 19

Fig. 20

Fig. 21

Fig. 22

Fig. 23

Fig. 24

Fig. 25

Fig. 26

Fig. 27

Fig. 28

Fig. 29

Fig. 30

Fig. 31

Fig. 32

Fig. 33

Fig. 34

Fig. 35

Fig. 36

Fig. 37

Fig. 38

Fig. 39

Fig. 40

Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments of the semiconductor device will be described with reference to the drawings. The embodiments shown below illustrate the configurations and methods for embodying the technical idea, and do not limit the materials, shapes, structures, arrangements, dimensions, etc. of each component to those described below. Various modifications can be made to the following embodiments.

[0013] (Configuration of the semiconductor device) With reference to FIGS. 1 to 9, an embodiment of the semiconductor device 10 will be described.

[0014] As shown in FIG. 1, the semiconductor device 10 is formed in a rectangular flat plate shape. The semiconductor device 10 has a device front surface 11 and a device back surface 12 facing opposite sides to each other, and a direction extends to intersecting both the device front surface 11 and the device back surface 12, the device side surfaces 13 to 16. In this embodiment, the device side surfaces 13 to 16 are facing a direction perpendicular to both the device front surface 11 and the device back surface 12.

[0015] The device front surface 11 and the device back surface 12 are arranged to be spaced apart from each other. In the following description, the arrangement direction between the device front surface 11 and the device back surface 12 is defined as the z direction. Note that the z direction can also be said to be the height direction of the semiconductor device 10. Of the directions perpendicular to the z direction, two mutually perpendicular directions are defined as the x direction and the y direction, respectively. The x direction is an example of the first direction, and the y direction is an example of the second direction.

[0016] In this embodiment, when viewed from the z - direction, the device side surfaces 13 and 14 are surfaces along the x - direction, and the device side surfaces 15 and 16 are surfaces along the y - direction. The device side surfaces 13 and 14 are surfaces facing opposite sides in the y - direction, and the device side surfaces 15 and 16 are surfaces facing opposite sides in the x - direction. In this embodiment, the shape of the semiconductor device 10 when viewed from the z - direction is a rectangular shape with the x - direction being the short - side direction and the y - direction being the long - side direction.

[0017] The semiconductor device 10 includes a support layer 20 and a resin layer 50 laminated on the support layer 20.

[0018] The support layer 20 is made of a material having electrical insulation properties, for example, an epoxy resin. The support layer 20 is formed in a rectangular flat plate shape with the z - direction as the thickness direction. Therefore, the thickness direction of the support layer 20 can also be said to be the height direction of the semiconductor device 10. The support layer 20 is disposed closer to the device back surface 12 than the device main surface 11 in the z - direction of the semiconductor device 10. The support layer 20 constitutes the device back surface 12 and a part of each of the z - directions of the device side surfaces 13 - 16.

[0019] The support layer 20 has a support main surface 21 and a support back surface 22 facing opposite sides in the z - direction, and a direction extends to orthogonal to both the support main surface 21 and the support back surface 22

[0020]

[0020] extends to support side surfaces 23 - 26. The support main surface 21 faces the same side as the device main surface 11, and the support back surface 22 faces the same side as the device back surface 12. In this embodiment, the support back surface 22 constitutes the device back surface 12. The support side surface 23 faces the same side as the device side surface 13, the support side surface 24 faces the same side as the device side surface 14, the support side surface 25 faces the same side as the device side surface 15, and the support side surface 26 faces the same side as the device side surface 16. The shape of the support layer 20 when viewed from the z - direction is a rectangular shape with the x - direction being the short - side direction and the y - direction being the long - side direction.As shown in FIG. 1, the resin layer 50 is formed in a rectangular flat plate shape with the z - direction as the thickness direction. Therefore, the thickness direction of the resin layer 50 can also be said to be the height direction of the semiconductor device 10. The resin layer 50 is formed on the support main surface 21 of the support layer 20. The resin layer 50 constitutes the device main surface 11 and a part of each of the device side surfaces 13 - 16 in the z - direction. The thickness of the resin layer 50 is thicker than the thickness of the support layer 20. The resin layer 50 is made of a material having electrical insulation properties. The resin layer 50 is made of, for example, the same material as the support layer 20. In this embodiment, the resin layer 50 is made of a black epoxy resin.

[0021] As shown in FIGS. 1 and 4, the resin layer 50 has a resin front surface 51 and a resin back surface 52 (see FIG. 5) facing opposite sides in the z - direction, and a direction extends to orthogonal to both the resin front surface 51 and the resin back surface 52, resin side surfaces 53 - 56. The resin front surface 51 faces the same side as the device main surface 11, and the resin back surface 52 faces the same side as the device back surface 12. In this embodiment, the resin front surface 51 constitutes the device main surface 11. The resin back surface 52 is in contact with the support main surface 21 of the support layer 20. The resin side surface 53 faces the same side as the device side surface 13, the resin side surface 54 faces the same side as the device side surface 14, the resin side surface 55 faces the same side as the device side surface 15, and the resin side surface 56 faces the same side as the device side surface 16. The shape of the resin layer 50 viewed from the z - direction is a rectangular shape with the y - direction as the long - side direction and the x - direction as the short - side direction. As shown in FIG. 1, in this embodiment, the resin side surface 53 and the support side surface 23 are flush, the resin side surface 54 and the support side surface 24 are flush, the resin side surface 55 and the support side surface 25 are flush, and the resin side surface 56 and the support side surface 26 are flush. Also, the device side surface 13 is constituted by the resin side surface 53 and the support side surface 23, the device side surface 14 is constituted by the resin side surface 54 and the support side surface 24, the device side surface 15 is constituted by the resin side surface 55 and the support side surface 25, and the device side surface 16 is constituted by the resin side surface 56 and the support side surface 26.

[0022] As shown in FIGS. 1 and 5, the resin layer 50 has an element encapsulation layer 50A formed on the support main surface 21 and a surface-side resin layer 50B laminated on the element encapsulation layer 50A. That is, the element encapsulation layer 50A and the surface-side resin layer 50B are laminated in the thickness direction (z direction) of the resin layer 50. The element encapsulation layer 50A constitutes the resin back surface 52 and a part of the resin side surfaces 53 to 56 in the z direction, and the surface-side resin layer 50B constitutes the resin main surface 51 and the remaining part of the resin side surfaces 53 to 56 in the z direction.

[0023] As shown in FIG. 5, the thickness of the surface-side resin layer 50B is thinner than the thickness of the element encapsulation layer 50A. The thickness of the surface-side resin layer 50B is preferably 1 / 2 or less of the thickness of the element encapsulation layer 50A. The thickness of the surface-side resin layer 50B may be such that it can encapsulate a plurality of wirings such as via connection wirings 63pc and 63qc described later formed on the element encapsulation layer 50A, and it is preferably as thin as possible within the range where these wirings can be encapsulated. An interface 57 is formed at the boundary between the element encapsulation layer 50A and the surface-side resin layer 50B. In the present embodiment, the interface 57 is formed as the xy plane.

[0024] As shown in FIG. 4, on the resin main surface 51 which is the surface of the resin layer 50, an external electrode 70 serving as an external terminal for electrically connecting to wiring or the like of a circuit board when the semiconductor device 10 is mounted on the circuit board, for example, is provided. That is, the resin main surface 51 becomes the mounting surface when the semiconductor device 10 is mounted on a circuit board, for example. The external electrode 70 includes a Cu (copper) layer and a plating layer covering the Cu layer. The Cu layer includes a seed layer formed on the resin main surface 51 and a plating layer formed on the seed layer. The seed layer is made of, for example, Cu or Ti (titanium). The plating layer is made of Cu. The plating layer covering the Cu layer is made of a laminate of, for example, a Ni (nickel) layer, a Pd (palladium) layer, and an Au (gold) layer. The external electrode 70 has a power supply electrode 71, a ground electrode 72, an output electrode 73, and a plurality of driver connection electrodes 74. The power supply electrode 71, the ground electrode 72, and the output electrode 73 are arranged closer to the resin side surface 54 than the plurality of driver connection electrodes 74 in the resin main surface 51 in the y direction. The power supply electrode 71, the ground electrode 72, and the output electrode 73 are arranged spaced apart from each other in the x direction in a state of being aligned with each other in the y direction. The plurality of driver connection electrodes 74 are arranged at both ends in the x direction of the resin main surface 51 and at the ends closer to the resin side surface 53 among both ends in the y direction of the resin main surface 51, respectively. Note that the number of the driver connection electrodes 74 can be arbitrarily changed.

[0025] Next, the internal structure of the semiconductor device 10 will be described.

[0026] As shown in FIGS. 2 and 3, the semiconductor device 10 includes a switching element 30 and a driver 40, and via conductors 60 electrically connected to the switching element 30 and the driver 40 individually. In the present embodiment, the switching element 30 includes a first switching element 30A and a second switching element 30B. That is, the semiconductor device 10 has a plurality of switching elements. The driver 40 includes a drive circuit for driving each of the plurality of switching elements. In the present embodiment, the driver 40 includes a drive circuit for driving each of the first switching element 30A and the second switching element 30B.

[0027] Each of the switching elements 30A, 30B and the driver 40 is mounted on the support layer 20. More specifically, as shown in FIGS. 5 and 7, each of the switching elements 30A, 30B and the driver 40 is joined onto the support main surface 21 of the support layer 20 by a joining material SD. The joining material SD is an adhesive such as an epoxy resin or a silicone resin, for example. Note that a joining material SD having good heat conductivity can also be used. Further, when a metal film is formed on the surfaces of the switching elements 30A, 30B and the driver 40 and the support layer 20 that face each other, an Ag (silver) paste, solder, or the like, or a conductive adhesive can also be used as the joining material SD. Further, the joining material SD may be omitted.

[0028] As shown in FIGS. 2 and 3, each of the switching elements 30A, 30B is, for example, a transistor. Each of the switching elements 30A, 30B is a transistor that operates at a high frequency of 1 MHz or more. In one example, each of the switching elements 30A, 30B is composed of GaN (Gallium Nitride). In the present embodiment, each of the switching elements 30A, 30B uses a gallium nitride high electron mobility transistor (GaNHEMT). Each of the switching elements 30A, 30B uses GaNHEMTs of the same size. In the present embodiment, each of the switching elements 30A, 30B uses GaNHEMTs having the same configuration. Each of the switching elements 30A, 30B is formed in a rectangular flat plate shape with the z direction as the thickness direction. For this reason, the thickness direction of each of the switching elements 30A, 30B can also be said to be the height direction of the semiconductor device 10. The shape of each of the switching elements 30A, 30B viewed from the z direction is a rectangular shape having a long side direction and a short side direction. In the present embodiment, each of the switching elements 30A, 30B is arranged on the support main surface 21 such that the y direction is the long side direction and the x direction is the short side direction.

[0029] Each of the switching elements 30A and 30B is arranged at an interval in a direction orthogonal to the height direction (z-direction) of the semiconductor device 10. As shown in FIG. 2, the switching elements 30A and 30B are arranged apart from each other in the x-direction while being aligned with each other in the y-direction. The first switching element 30A is arranged closer to the support side surface 25 than the second switching element 30B on the support main surface 21. Thus, the x-direction can also be said to be the arrangement direction of the first switching element 30A and the second switching element 30B.

[0030] As shown in FIG. 5, the first switching element 30A has a first element main surface 31A and a first element back surface 32A facing opposite sides in the z-direction. The first element main surface 31A faces the same side as the device main surface 11 (support main surface 21), and the first element back surface 32A faces the same side as the device back surface 12 (support back surface 22).

[0031] As shown in FIG. 2, a first drive pad electrode 31AA, a second drive pad electrode 31AB, and a control pad electrode 31AC are formed on the first element main surface 31A. In the present embodiment, the first drive pad electrode 31AA constitutes the drain electrode of the first switching element 30A, the second drive pad electrode 31AB constitutes the source electrode of the first switching element 30A, and the control pad electrode 31AC constitutes the gate electrode of the first switching element 30A. Thus, the drive pad electrodes 31AA and 31AB and the control pad electrode 31AC are formed on the surface of the first switching element 30A closer to the exterior electrode 70 in the z-direction. For this reason, the distances between the drive pad electrodes 31AA and 31AB and the power supply electrode 71 and the output electrode 73 are shortened.

[0032] Each drive pad electrode 31AA, 31AB has the same size and is rectangular with a long side direction and a short side direction when viewed from the z direction. Each drive pad electrode 31AA, 31AB is formed on the first element main surface 31A such that the long side direction is along the x direction and the short side direction is along the y direction. That is, each drive pad electrode 31AA, 31AB extends in the x direction, which is the arrangement direction of the first switching element 30A and the second switching element 30B. As shown in FIG. 2, a plurality of first drive pad electrodes 31AA and a plurality of second drive pad electrodes 31AB are formed on the first element main surface 31A. The first drive pad electrodes 31AA and the second drive pad electrodes 31AB are alternately arranged in the y direction. The control pad electrode 31AC is arranged at an end of the first element main surface 31A in the y direction that is closer to the support side surface 23 (driver 40) among the two end portions. That is, the control pad electrode 31AC is arranged closer to the support side surface 23 (driver 40) than each drive pad electrode 31AA, 31AB on the first element main surface 31A.

[0033] As shown in FIG. 5, the second switching element 30B has a second element main surface 31B and a second element back surface 32B that face opposite sides in the z direction. The second element main surface 31B faces the same side as the device main surface 11 (support main surface 21), and the second element back surface 32B faces the same side as the device back surface 12 (support back surface 22).

[0034] As shown in FIG. 2, a first drive pad electrode 31BA, a second drive pad electrode 31BB, and a control pad electrode 31BC are formed on the second element main surface 31B. In the present embodiment, the first drive pad electrode 31BA constitutes the drain electrode of the second switching element 30B, the second drive pad electrode 31BB constitutes the source electrode of the second switching element 30B, and the control pad electrode 31BC constitutes the gate electrode of the second switching element 30B. Thus, each drive pad electrode 31BA, 31BB and the control pad electrode 31BC are formed on the surface of the second switching element 30B that is closer to the exterior electrode 70 in the z direction. For this reason, the distance between each drive pad electrode 31BA, 31BB and the ground electrode 72 and the output electrode 73 is shortened.

[0035] Each of the drive pad electrodes 31BA and 31BB has the same size and is rectangular with a long side direction and a short side direction when viewed from the z direction. Each of the drive pad electrodes 31BA and 31BB is formed on the second element main surface 31B such that the long side direction is along the x direction and the short side direction is along the y direction. That is, each of the drive pad electrodes 31BA and 31BB extends in the x direction, which is the arrangement direction of the first switching element 30A and the second switching element 30B. As shown in FIG. 2, a plurality of first drive pad electrodes 31BA and a plurality of second drive pad electrodes 31BB are formed on the second element main surface 31B. The first drive pad electrodes 31BA and the second drive pad electrodes 31BB are alternately arranged in the y direction. When viewed from the z direction, the first drive pad electrode 31BA is arranged at a position aligned in the y direction with the first drive pad electrode 31AA of the first switching element 30A, and the second drive pad electrode 31BB is arranged at a position aligned in the y direction with the second drive pad electrode 31AB of the first switching element 30A. In other words, when viewed from the z direction, the first drive pad electrode 31BA is arranged at a position shifted in the y direction with respect to the second drive pad electrode 31AB, and the second drive pad electrode 31BB is arranged at a position shifted in the y direction with respect to the first drive pad electrode 31AA. The control pad electrode 31BC is arranged at an end portion closer to the support side surface 23 (driver 40) among both end portions in the y direction of the second element main surface 31B. That is, the control pad electrode 31BC is arranged closer to the support side surface 23 (driver 40) than each of the drive pad electrodes 31BA and 31BB on the second element main surface 31B.

[0036] Thus, each of the switching elements 30A and 30B is a lateral transistor in which a first drive electrode (drain electrode), a second drive electrode (source electrode), and a control electrode (gate electrode) are formed on one surface facing the z direction.

[0037] Note that each of the switching elements 30A and 30B is not limited to a gallium nitride high electron mobility transistor, and may be a MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) containing Si (silicon) or a MOSFET containing SiC (silicon carbide).

[0038] As shown in FIGS. 2 and 4, the driver 40 is arranged at a distance from each of the switching elements 30A and 30B in the y direction. When viewed from the z direction, the driver 40 is arranged closer to the support side surface 23 than each of the switching elements 30A and 30B in the y direction on the support main surface 21. In the present embodiment, the driver 40 is arranged at the center in the x direction of the support main surface 21. As shown in FIG. 7, the driver 40 is arranged at a position aligned with each of the switching elements 30A and 30B in the z direction. That is, each of the switching elements 30A and 30B and the driver 40 are arranged on the same plane.

[0039] As shown in FIGS. 1 and 2, the driver 40 is formed in a rectangular flat plate shape with the z direction as the thickness direction. Therefore, the thickness direction of the driver 40 can also be said to be the height direction of the semiconductor device 10. In the present embodiment, the shape of the driver 40 when viewed from the z direction is a rectangular shape in which the x direction is the long side direction and the y direction is the short side direction. Note that the shape of the driver 40 when viewed from the z direction can be arbitrarily changed. In one example, the shape of the driver 40 when viewed from the z direction may be a square.

[0040] As shown in Fig. 7, the driver 40 has a driver front surface 41 and a driver back surface 42 that face opposite sides in the z direction. The driver front surface 41 faces the same side as the device front surface 11, and the driver back surface 42 faces the same side as the device back surface 12. Also, the driver 40 has a plurality of driver pad electrodes 43. Each driver pad electrode 43 is exposed from the driver front surface 41. That is, each driver pad electrode 43 is exposed from the driver front surface 41 of the driver 40, which faces the same side as each element front surface 31A, 31B on which the control pad electrodes 31AC, 31BC are formed. Therefore, in the z direction, the variation in the positions of each driver pad electrode 43 and the positions of each control pad electrode 31AC, 31BC is small. Also, each driver pad electrode 43 is exposed from the surface of the driver 40 that is closer to the exterior electrode 70 in the z direction. For this reason, in the z direction, the distance between the driver pad electrode 43 and the driver connection electrode 74 becomes short.

[0041] As shown in Figs. 5 and 7, each of the switching elements 30A, 30B and the driver 40 is encapsulated by a resin layer 50. More specifically, each of the switching elements 30A, 30B and the driver 40 is encapsulated by an element encapsulation layer 50A. The interface 57 between the element encapsulation layer 50A and the surface-side resin layer 50B is formed closer to the resin front surface 51 (device front surface 11) than each element front surface 31A, 31B of each switching element 30A, 30B and the driver front surface 41 of the driver 40. That is, the element encapsulation layer 50A covers each element front surface 31A, 31B of each switching element 30A, 30B and the driver front surface 41 of the driver 40.

[0042] As shown in FIG. 4, when viewed from the z direction, the first switching element 30A is disposed at a position overlapping with the power supply electrode 71. In other words, the power supply electrode 71 is disposed at a position overlapping with the first switching element 30A when viewed from the z direction. More specifically, the power supply electrode 71 is disposed at a position overlapping with each first drive pad electrode 31AA of the first switching element 30A when viewed from the z direction. The power supply electrode 71 is disposed so as to protrude in the x direction with respect to the first switching element 30A. As shown in FIG. 4, the power supply electrode 71 has a protruding portion that protrudes toward the device side surface 15 with respect to the first switching element 30A.

[0043] When viewed from the z direction, the second switching element 30B is disposed at a position overlapping with the ground electrode 72. In other words, the ground electrode 72 is disposed at a position overlapping with the second switching element 30B when viewed from the z direction. More specifically, the ground electrode 72 is disposed at a position overlapping with each second drive pad electrode 31BB of the second switching element 30B when viewed from the z direction. The ground electrode 72 is disposed so as to protrude in the x direction with respect to the second switching element 30B. As shown in FIG. 4, the ground electrode 72 has a protruding portion that protrudes toward the device side surface 16 with respect to the second switching element 30B.

[0044] When viewed from the z direction, the first switching element 30A and the second switching element 30B are disposed at positions not overlapping with the output electrode 73. In other words, the output electrode 73 is disposed at a position not overlapping with both the first switching element 30A and the second switching element 30B. In the present embodiment, the output electrode 73 is disposed between the first switching element 30A and the second switching element 30B in the x direction when viewed from the z direction.

[0045] When viewed from the z direction, the driver 40 is disposed at a position that does not overlap with the plurality of driver connection electrodes 74. In other words, the plurality of driver connection electrodes 74 are disposed at positions that do not overlap with the driver 40 when viewed from the z direction. More specifically, the driver connection electrodes 74 are disposed so as to surround the driver 40 when viewed from the z direction.

[0046] As shown in FIGS. 3 and 5 to 8, the via conductor 60 is provided in the resin layer 50 and is a conductor that electrically connects the exterior electrode 70 and each of the switching elements 30A and 30B, electrically connects each of the switching elements 30A and 30B and the driver 40, or electrically connects the driver 40 and the exterior electrode 70. The via conductor 60 includes a power supply via conductor 61, a ground via conductor 62, an output via conductor 63, an element control via conductor 64, and a plurality of driver via conductors 65. The via conductor 60 is made of, for example, Cu.

[0047] As shown in FIGS. 3 and 6, the power supply via conductor 61 is a conductor that electrically connects the first drive pad electrode 31AA of the first switching element 30A and the power supply electrode 71 of the exterior electrode 70. In the present embodiment, a plurality of power supply via conductors 61 are provided. The plurality of power supply via conductors 61 are arranged at intervals in the y direction while being aligned with each other in the x direction.

[0048] As shown in FIG. 3, the power supply via conductor 61 is disposed at a position that overlaps with the first drive pad electrode 31AA of the first switching element 30A when viewed from the z direction. In the present embodiment, the power supply via conductor 61 is disposed at a position that overlaps with the end portion of the first drive pad electrode 31AA closer to the support side surface 25 among the both end portions in the x direction. In other words, the power supply via conductor 61 is disposed at the end portion of the first drive pad electrode 31AA farther from the output electrode 73 among the both end portions in the x direction.

[0049] As shown in FIG. 4, the power supply via conductor 61 is arranged at a position overlapping with the power supply electrode 71 when viewed from the z direction. That is, as shown in FIG. 6, the power supply via conductor 61 is arranged at a position overlapping with both the first drive pad electrode 31AA and the power supply electrode 71 when viewed from the z direction. Each power supply via conductor 61 is composed of a via extending along the z direction. In the present embodiment, the shape of each power supply via conductor 61 when viewed from the z direction is circular. The diameter of each power supply via conductor 61 is, for example, 100 μm or more and 200 μm or less. Note that the shape of the power supply via conductor 61 when viewed from the z direction is not limited to a circle and can be arbitrarily changed. For example, the shape of the power supply via conductor 61 when viewed from the z direction may be a polygon such as a quadrilateral or an ellipse.

[0050] As shown in FIG. 6, the power supply via conductor 61 is provided so as to penetrate the resin layer 50 in the z direction. In other words, the power supply via conductor 61 is provided so as to penetrate both the element sealing layer 50A and the surface-side resin layer 50B in the z direction. More specifically, the resin layer 50 is provided with a through hole 58a that penetrates a portion in the z direction between the first element main surface 31A of the first switching element 30A and the interface 57 in the element sealing layer 50A and the entire z direction of the surface-side resin layer 50B. The first drive pad electrode 31AA is exposed in the z direction through the through hole 58a. The through hole 58a exposes, in the z direction, the end portion closer to the resin side surface 55 among both end portions in the x direction of the first drive pad electrode 31AA. The power supply via conductor 61 is provided so as to fill the through hole 58a. For this reason, the power supply via conductor 61 is in contact with the first drive pad electrode 31AA and is exposed from the resin main surface 51 of the resin layer 50. In other words, the power supply via conductor 61 is connected to the end portion (the end portion closer to the resin side surface 55) among both end portions in the x direction of the first drive pad electrode 31AA that is farther from the output electrode 73. And the power supply via conductor 61 is exposed from a position overlapping the end portion (the end portion closer to the resin side surface 55) among both end portions in the x direction of the first drive pad electrode 31AA that is farther from the output electrode 73 when viewed in the z direction on the resin main surface 51 of the resin layer 50. The portion of the power supply via conductor 61 exposed from the resin main surface 51 is covered by the power supply electrode 71. That is, the power supply via conductor 61 is in contact with the power supply electrode 71. Thereby, the power supply via conductor 61 is connected to the first drive pad electrode 31AA and the power supply electrode 71.

[0051] In the present embodiment, the length of the power supply via conductor 61 in the z direction is slightly longer than the length of the first switching element 30A in the z direction (the thickness of the first switching element 30A) and shorter than the thickness of the element sealing layer 50A.

[0052] As shown in FIGS. 3 and 6, the ground via conductor 62 is a conductor that electrically connects the second drive pad electrode 31BB of the second switching element 30B and the ground electrode 72 among the exterior electrodes 70. In the present embodiment, a plurality of ground via conductors 62 are provided. The plurality of ground via conductors 62 are arranged at intervals in the y direction in a state of being aligned with each other in the x direction. When viewed from the z direction, the ground via conductor 62 is arranged so as to be displaced from the power supply via conductor 61 in the y direction.

[0053] As shown in FIG. 3, the ground via conductor 62 is arranged at a position overlapping the second drive pad electrode 31BB of the second switching element 30B when viewed from the z direction. In the present embodiment, the ground via conductor 62 is arranged at a position overlapping the end portion closer to the support side surface 26 among both end portions in the x direction of the second drive pad electrode 31BB. In other words, the ground via conductor 62 is arranged at the end portion farther from the output electrode 73 among both end portions in the x direction of the second drive pad electrode 31BB.

[0054] As shown in FIG. 4, the ground via conductor 62 is arranged at a position overlapping the ground electrode 72 when viewed from the z direction. That is, as shown in FIG. 6, the ground via conductor 62 is arranged at a position overlapping both the second drive pad electrode 31BB and the ground electrode 72 when viewed from the z direction. Each ground via conductor 62 is composed of a via extending along the z direction. In the present embodiment, the shape of each ground via conductor 62 when viewed from the z direction is circular. The diameter of each ground via conductor 62 is, for example, 100 μm or more and 200 μm or less. Note that the shape of the ground via conductor 62 when viewed from the z direction is not limited to a circle and can be arbitrarily changed. For example, the shape of the ground via conductor 62 when viewed from the z direction may be a polygon such as a quadrangle or an ellipse.

[0055] As shown in FIG. 6, the ground via conductor 62 is provided so as to penetrate the resin layer 50 in the z direction. In other words, the ground via conductor 62 is provided so as to penetrate both the element encapsulation layer 50A and the surface-side resin layer 50B in the z direction. More specifically, the resin layer 50 is provided with a through hole 58b that penetrates a portion in the z direction between the second element main surface 31B of the second switching element 30B and the interface 57 in the element encapsulation layer 50A and the entire z direction of the surface-side resin layer 50B. The second drive pad electrode 31BB is exposed in the z direction through the through hole 58b. The through hole 58b exposes, in the z direction, an end portion closer to the support side surface 26 among both end portions of the second drive pad electrode 31BB in the x direction. The ground via conductor 62 is provided so as to fill the through hole 58b. For this reason, the ground via conductor 62 is in contact with the second drive pad electrode 31BB and is exposed from the resin main surface 51 of the resin layer 50. In other words, the ground via conductor 62 is connected to an end portion (the end portion closer to the support side surface 26) of both end portions of the second drive pad electrode 31BB in the x direction that is farther from the output electrode 73. And the ground via conductor 62 is exposed from a position overlapping an end portion (the end portion closer to the support side surface 26) of both end portions of the second drive pad electrode 31BB in the x direction that is farther from the output electrode 73 as viewed in the z direction from the resin main surface 51 of the resin layer 50. The portion of the ground via conductor 62 exposed from the resin main surface 51 is covered by the ground electrode 72. That is, the ground via conductor 62 is in contact with the ground electrode 72. Thereby, the ground via conductor 62 is connected to the second drive pad electrode 31BB and the ground electrode 72.

[0056] In this embodiment, the length of the ground via conductor 62 in the z direction is slightly longer than the length of the second switching element 30B in the z direction (the thickness of the second switching element 30B), and shorter than the thickness of the element sealing layer 50A. The length of the ground via conductor 62 in the z direction is equal to the length of the power supply via conductor 61 in the z direction. Here, if the difference between the length of the ground via conductor 62 in the z direction and the length of the power supply via conductor 61 in the z direction is, for example, 10% or less of the length of the ground via conductor 62 in the z direction, it can be said that the length of the ground via conductor 62 in the z direction is equal to the length of the power supply via conductor 61 in the z direction.

[0057] As shown in FIGS. 3 and 5, the output via conductor 63 is a conductor that electrically connects the second drive pad electrode 31AB of the first switching element 30A, the first drive pad electrode 31BA of the second switching element 30B, and the output electrode 73 among the exterior electrodes 70.

[0058] The output via conductor 63 includes a first output via conductor 63P that connects the second drive pad electrode 31AB and the output electrode 73, and a second output via conductor 63Q that connects the first drive pad electrode 31BA and the output electrode 73. The first output via conductor 63P and the second output via conductor 63Q are electrically connected via the output electrode 73. In other words, the first output via conductor 63P and the second output via conductor 63Q are not directly electrically connected. When viewed from the z direction, since the second drive pad electrode 31AB and the first drive pad electrode 31BA are displaced from each other in the y direction, the first output via conductor 63P and the second output via conductor 63Q are displaced from each other in the y direction. Of the two end portions of the first output via conductor 63P in the y direction, the end portion closer to the resin side surface 53 is disposed closer to the resin side surface 53 than the second output via conductor 63Q. In other words, of the two end portions of the second output via conductor 63Q in the y direction, the end portion closer to the resin side surface 54 is disposed closer to the resin side surface 54 than the first output via conductor 63P.

[0059] As shown in FIG. 5, in the x direction, the first output via conductor 63P is disposed closer to the resin side surface 55 than the second output via conductor 63Q. The space between the first output via conductor 63P and the second output via conductor 63Q in the x direction is filled with the resin layer 50.

[0060] The first output via conductor 63P is provided so as to penetrate the resin layer 50 in the z direction. More specifically, the first output via conductor 63P is provided so as to penetrate both the portion of the element sealing layer 50A that covers the first element main surface 31A of the first switching element 30A and the surface-side resin layer 50B in the z direction. As shown in FIG. 5, the first output via conductor 63P is in a crank shape. The first output via conductor 63P includes a first element connection via 63pa connected to the second drive pad electrode 31AB, a first electrode connection via 63pb connected to the output electrode 73, and a first via connection wiring 63pc that connects the first element connection via 63pa and the first electrode connection via 63pb.

[0061] As shown in FIG. 3, in the present embodiment, a plurality of first element connection vias 63pa are provided. The plurality of first element connection vias 63pa are arranged at intervals in the y direction while being aligned with each other in the x direction. When viewed from the z direction, the first element connection via 63pa is disposed at a position overlapping the second drive pad electrode 31AB. More specifically, when viewed from the z direction, the first element connection via 63pa is disposed at a position overlapping the end portion of the second drive pad electrode 31AB closer to the support side surface 26 among the both end portions in the x direction. In other words, the first element connection via 63pa is disposed at a position overlapping the end portion of the second drive pad electrode 31AB closer to the output electrode 73 among the both end portions in the x direction. That is, the first element connection via 63pa is disposed closer to the support side surface 26 (closer to the output electrode 73) than the power supply via conductor 61 in the x direction. When viewed from the z direction, the plurality of first element connection vias 63pa are disposed at positions shifted from the plurality of power supply via conductors 61 in the y direction. Also, when viewed from the z direction, the plurality of first element connection vias 63pa are disposed at aligned positions with respect to the plurality of ground via conductors 62 in the y direction.

[0062] As shown in FIG. 5, the first element connection via 63pa penetrates in the z direction through a portion of the element sealing layer 50A that covers the first element main surface 31A. More specifically, the element sealing layer 50A is provided with a through hole 58c that penetrates in the z direction through a portion of the element sealing layer 50A that covers the first element main surface 31A. The second drive pad electrode 31AB is exposed in the z direction through the through hole 58c. The through hole 58c exposes, in the z direction, an end portion (the end portion closer to the output electrode 73) of both ends of the second drive pad electrode 31AB in the x direction that is closer to the support side surface 26. The first element connection via 63pa is provided so as to fill the through hole 58c. For this reason, the first element connection via 63pa extends along the z direction and is in contact with the second drive pad electrode 31AB. In other words, the first element connection via 63pa is connected to an end portion (the end portion closer to the resin side surface 56) of both ends of the second drive pad electrode 31AB in the x direction that is closer to the output electrode 73. And the first element connection via 63pa is exposed from a position that overlaps with an end portion (the end portion closer to the resin side surface 56) of both ends of the second drive pad electrode 31AB in the x direction when viewed from the z direction in the element sealing layer 50A.

[0063] As shown in FIG. 3, the shape of the first element connection via 63pa when viewed from the z direction is circular. The diameter of the first element connection via 63pa is equal to the diameter of the power supply via conductor 61, for example, 100 μm or more and 200 μm or less. Note that the shape of the first element connection via 63pa when viewed from the z direction is not limited to circular and can be arbitrarily changed. For example, the shape of the first element connection via 63pa when viewed from the z direction may be a polygon such as a quadrilateral or an ellipse. Also, in the present embodiment, the length of the first element connection via 63pa in the z direction is shorter than the length of the first switching element 30A in the z direction. The length of the first element connection via 63pa in the z direction is less than 1 mm. The length of the first element connection via 63pa in the z direction is about several hundred μm.

[0064] As shown in FIG. 5, the first via connection wiring 63pc is provided on the element encapsulation layer 50A. In other words, the first via connection wiring 63pc is provided at the interface 57 between the element encapsulation layer 50A and the surface-side resin layer 50B. As shown in FIG. 3, the shape of the first via connection wiring 63pc viewed from the z direction is a rectangular shape with the x direction being the short side direction and the y direction being the long side direction. Viewed from the z direction, the first via connection wiring 63pc is formed so as to straddle each of the second drive pad electrodes 31AB in the y direction. Viewed from the z direction, the first via connection wiring 63pc covers each of the first element connection vias 63pa. In other words, each of the first element connection vias 63pa is arranged at a position overlapping the first via connection wiring 63pc when viewed from the z direction. More specifically, the first element connection via 63pa is arranged at a position overlapping both the second drive pad electrode 31AB and the first via connection wiring 63pc when viewed from the z direction.

[0065] In the present embodiment, the first element connection via 63pa is arranged at a position overlapping the end portion (the end portion farther from the output electrode 73) closer to the resin side surface 55 among the both end portions in the x direction of the first via connection wiring 63pc. Also, the end portion closer to the resin side surface 55 among the both end portions in the x direction of the first via connection wiring 63pc is a portion protruding in the x direction from the output electrode 73 when viewed from the z direction. That is, as shown in FIG. 5, when viewed from the z direction, the first element connection via 63pa is arranged at a position not overlapping the output electrode 73. More specifically, the first element connection via 63pa is arranged closer to the resin side surface 55 than the output electrode 73 in the x direction.

[0066] As shown in FIG. 3, the first via connection wiring 63pc is arranged at a position shifted in the x direction with respect to the first switching element 30A when viewed from the z direction. More specifically, the first via connection wiring 63pc has an overhanging portion that protrudes toward the resin side surface 56 in the x direction with respect to the first switching element 30A when viewed from the z direction. As shown in FIG. 5, the overhanging portion of the first via connection wiring 63pc extends to a position overlapping the output electrode 73 when viewed from the z direction. In the present embodiment, when viewed from the z direction, the center of the first via connection wiring 63pc in the y direction is arranged at a position aligned in the y direction with the center of the first switching element 30A in the y direction.

[0067] As shown in FIG. 3, in the present embodiment, a plurality of first electrode connection vias 63pb are provided. The plurality of first electrode connection vias 63pb are arranged at intervals from each other in the y direction in a state where they are aligned with each other in the x direction. In the present embodiment, the first electrode connection vias 63pb are arranged at positions aligned with the first element connection vias 63pa in the y direction. That is, both the first electrode connection vias 63pb and the first element connection vias 63pa are arranged at positions shifted with respect to the power supply via conductor 61 in the y direction. Note that the arrangement position of the first electrode connection vias 63pb in the y direction can be arbitrarily changed. In one example, the first electrode connection vias 63pb may be arranged at positions shifted with respect to the first element connection vias 63pa in the y direction. At least one of the plurality of first electrode connection vias 63pb may be arranged at a position aligned with the power supply via conductor 61 in the y direction.

[0068] As shown in FIG. 4, when viewed from the z direction, the via 63pb for connecting the first electrode is disposed at a position overlapping the output electrode 73. More specifically, when viewed from the z direction, the via 63pb for connecting the first electrode is disposed at a position overlapping the end portion closer to the support side surface 25 among the both end portions in the x direction of the output electrode 73. Also, when viewed from the z direction, the via 63pb for connecting the first electrode is disposed at a position overlapping the wiring 63pc for connecting the first via. In the present embodiment, the via 63pb for connecting the first electrode is disposed at a position overlapping the end portion closer to the resin side surface 56 (the end portion closer to the output electrode 73) among the both end portions in the x direction of the wiring 63pc for connecting the first via. Further, as shown in FIG. 5, the end portion closer to the resin side surface 56 among the both end portions in the x direction of the wiring 63pc for connecting the first via is a portion overlapping the output electrode 73 when viewed from the z direction. In other words, when viewed from the z direction, the via 63pb for connecting the first electrode is disposed at a position overlapping the protruding portion of the wiring 63pc for connecting the first via. Thus, when viewed from the z direction, the via 63pb for connecting the first electrode is disposed at a position overlapping the output electrode 73 and not overlapping the first switching element 30A. When viewed from the z direction, the via 63pb for connecting the first electrode is disposed between the first switching element 30A and the second switching element 30B in the x direction.

[0069] As shown in FIG. 5, the via 63pb for connecting the first electrode is disposed at a position shifted from the via 63pa for connecting the first element in the z direction. The via 63pb for connecting the first electrode is provided so as to penetrate the surface-side resin layer 50B in the z direction. More specifically, the resin layer 50 is provided with a through hole 58d that penetrates the surface-side resin layer 50B in the z direction. Through the through hole 58d, the wiring 63pc for connecting the first via is exposed in the z direction. The through hole 58d exposes, in the z direction, the end portion (the end portion closer to the output electrode 73) of the wiring 63pc for connecting the first via that is closer to the support side surface 26 among both end portions in the x direction of the wiring 63pc for connecting the first via. The via 63pb for connecting the first electrode is provided so as to fill the through hole 58d. For this reason, the via 63pb for connecting the first electrode extends along the z direction and is in contact with the wiring 63pc for connecting the first via. Further, the via 63pb for connecting the first electrode is exposed from the resin layer 50 in the z direction. In other words, the via 63pb for connecting the first electrode is connected to the end portion (the end portion closer to the resin side surface 56) of the wiring 63pc for connecting the first via that is closer to the output electrode 73 among both end portions in the x direction of the wiring 63pc for connecting the first via. And the via 63pb for connecting the first electrode is exposed from a position overlapping the end portion (the end portion closer to the resin side surface 56) of the wiring 63pc for connecting the first via that is closer to the output electrode 73 among both end portions in the x direction when viewed from the z direction on the resin main surface 51 of the resin layer 50. The portion of the via 63pb for connecting the first electrode that is exposed from the resin layer 50 is covered by the output electrode 73. For this reason, the via 63pb for connecting the first electrode is in contact with the output electrode 73. That is, the via 63pb for connecting the first electrode connects both the wiring 63pc for connecting the first via and the output electrode 73.

[0070] As shown in FIG. 3, the shape of the first electrode connection via 63pb viewed from the z direction is circular. The diameter of the first electrode connection via 63pb is equal to the diameter of the first element connection via 63pa, for example, 100 μm or more and 200 μm or less. Note that the shape of the first electrode connection via 63pb viewed from the z direction is not limited to a circle and can be arbitrarily changed. For example, the shape of the first electrode connection via 63pb viewed from the z direction may be a polygon such as a quadrilateral or an ellipse. In the present embodiment, the length of the first electrode connection via 63pb in the z direction is longer than the length of the first element connection via 63pa in the z direction. Note that the length of the first electrode connection via 63pb in the z direction can be arbitrarily changed. In one example, the length of the first electrode connection via 63pb in the z direction may be equal to the length of the first element connection via 63pa in the z direction. In this case, the thickness of the portion of the element sealing layer 50A that covers the first element main surface 31A of the first switching element 30A is equal to the thickness of the surface-side resin layer 50B.

[0071] As shown in FIG. 5, the second output via conductor 63Q is provided so as to penetrate the resin layer 50 in the z direction. More specifically, the second output via conductor 63Q is provided so as to penetrate both the portion of the element sealing layer 50A that covers the second element main surface 31B of the second switching element 30B and the surface-side resin layer 50B in the z direction. As shown in FIG. 5, the second output via conductor 63Q has a crank shape. The second output via conductor 63Q has a symmetrical shape with the first output via conductor 63P. The second output via conductor 63Q includes a second element connection via 63qa connected to the first drive pad electrode 31BA, a second electrode connection via 63qb connected to the output electrode 73, and a second via connection wiring 63qc that connects the second element connection via 63qa and the second electrode connection via 63qb.

[0072] As shown in FIG. 3, in the present embodiment, a plurality of vias 63qa for connecting the second element are provided. The plurality of vias 63qa for connecting the second element are arranged at intervals in the y direction while being aligned with each other in the x direction. When viewed from the z direction, the vias 63qa for connecting the second element are arranged at positions overlapping the first drive pad electrode 31BA. More specifically, when viewed from the z direction, the vias 63qa for connecting the second element are arranged at positions overlapping the end portion closer to the resin side surface 55 among the both end portions in the x direction of the first drive pad electrode 31BA. That is, the vias 63qa for connecting the second element are arranged closer to the resin side surface 55 than the ground via conductor 62 in the x direction. The vias 63qa for connecting the second element are arranged at positions shifted from the ground via conductor 62 in the y direction. The vias 63qa for connecting the second element are arranged at positions shifted from the vias 63pa for connecting the first element and the vias 63pb for connecting the first electrode in the y direction. In the present embodiment, the vias 63qa for connecting the second element are arranged at positions aligned with the power supply via conductor 61 in the y direction.

[0073] As shown in FIG. 5, the vias 63qa for connecting the second element penetrate in the z direction through the portion of the element sealing layer 50A that covers the second element main surface 31B. More specifically, the element sealing layer 50A is provided with a through hole 58e that penetrates in the z direction through the portion of the element sealing layer 50A that covers the second element main surface 31B. The first drive pad electrode 31BA is exposed in the z direction through the through hole 58e. The vias 63qa for connecting the second element are provided so as to fill the through hole 58e. For this reason, the vias 63qa for connecting the second element extend along the z direction and are in contact with the first drive pad electrode 31BA. In other words, the vias 63qa for connecting the second element are connected to the end portion closer to the output electrode 73 (the end portion closer to the resin side surface 55) among the both end portions in the x direction of the first drive pad electrode 31BA. Then, the vias 63qa for connecting the second element are exposed from the position overlapping the end portion closer to the output electrode 73 (the end portion closer to the resin side surface 55) among the both end portions in the x direction of the first drive pad electrode 31BA when viewed from the z direction in the element sealing layer 50A.

[0074] As shown in FIG. 3, the shape of the via 63qa for connecting the second element viewed from the z direction is circular. In the present embodiment, the diameter of the via 63qa for connecting the second element is equal to the diameter of the ground via conductor 62, for example, 100 μm or more and 200 μm or less. Note that the shape of the via 63qa for connecting the second element viewed from the z direction is not limited to a circle and can be arbitrarily changed. For example, the shape of the via 63qa for connecting the second element viewed from the z direction may be a polygon such as a quadrilateral or an ellipse. Also, in the present embodiment, the length of the via 63qa for connecting the second element in the z direction is shorter than the length of the second switching element 30B in the z direction. The length of the via 63qa for connecting the second element in the z direction is less than 1 mm. The length of the via 63qa for connecting the second element in the z direction is about several hundred μm. Thus, in the present embodiment, the number, shape, and size of the via 63qa for connecting the second element are the same as the number, shape, and size of the via 63pa for connecting the first element.

[0075] As shown in FIG. 5, the wiring 63qc for the second via connection is provided on the element encapsulation layer 50A. In other words, the wiring 63qc for the second via connection is provided at the interface 57 between the element encapsulation layer 50A and the surface-side resin layer 50B. As shown in FIG. 3, the shape of the wiring 63qc for the second via connection viewed from the z direction is a rectangular shape with the x direction being the short side direction and the y direction being the long side direction. Viewed from the z direction, the wiring 63qc for the second via connection is formed so as to straddle each of the first drive pad electrodes 31BA in the y direction. Viewed from the z direction, the wiring 63qc for the second via connection covers each of the second element connection vias 63qa. In other words, each of the second element connection vias 63qa is arranged at a position overlapping with the wiring 63qc for the second via connection when viewed from the z direction. More specifically, the second element connection via 63qa is arranged at a position overlapping both the first drive pad electrode 31BA and the wiring 63qc for the second via connection when viewed from the z direction. In the present embodiment, the shape and size of the wiring 63qc for the second via connection are the same as those of the wiring 63pc for the first via connection. Also, the wiring 63qc for the second via connection is arranged offset from the wiring 63pc for the first via connection in the y direction. More specifically, as shown in FIG. 3, the center of the wiring 63qc for the second via connection in the y direction is located closer to the resin side surface 54 (device side surface 14) than the center of the wiring 63pc for the first via connection in the y direction. When viewed from the x direction, the wiring 63qc for the second via connection overlaps with the wiring 63pc for the first via connection. That is, the end of the wiring 63qc for the second via connection closer to the resin side surface 54 (device side surface 14) among both ends in the y direction is located closer to the resin side surface 54 (device side surface 14) than the wiring 63pc for the first via connection. In other words, the end of the wiring 63pc for the first via connection closer to the resin side surface 53 (device side surface 13) among both ends in the y direction is located closer to the resin side surface 53 (device side surface 13) than the wiring 63qc for the second via connection.

[0076] As shown in FIG. 3, the center of the second via connection wiring 63qc in the y direction is arranged at a position deviated in the y direction with respect to the center of the second switching element 30B in the y direction. Specifically, the center of the second via connection wiring 63qc in the y direction is located closer to the resin side surface 54 than the center of the second switching element 30B in the y direction. The second via connection wiring 63qc is arranged at a position deviated in the x direction with respect to the second switching element 30B when viewed from the z direction. More specifically, the second via connection wiring 63qc has an overhanging portion that protrudes toward the support side surface 25 in the x direction with respect to the second switching element 30B when viewed from the z direction. As shown in FIG. 5, the overhanging portion of the second via connection wiring 63qc extends to a position overlapping the output electrode 73 when viewed from the z direction. The second via connection wiring 63qc is arranged at a distance from the first via connection wiring 63pc in the x direction.

[0077] In the present embodiment, the second element connection via 63qa is arranged at a position overlapping the end portion closer to the resin side surface 55 (the end portion closer to the output electrode 73) among the both end portions of the second via connection wiring 63qc in the x direction. Also, the end portion closer to the resin side surface 56 among the both end portions of the second via connection wiring 63qc in the x direction is a portion protruding in the x direction from the output electrode 73 when viewed from the z direction. That is, when viewed from the z direction, the second element connection via 63qa is arranged at a position not overlapping the output electrode 73. The end portion closer to the resin side surface 56 among the both end portions of the second via connection wiring 63qc in the x direction is arranged at a position overlapping the second switching element 30B when viewed from the z direction. For this reason, the second element connection via 63qa is arranged closer to the resin side surface 56 than the output electrode 73 in the x direction.

[0078] As shown in FIG. 3, in the present embodiment, a plurality of vias 63qb for connecting the second electrode are provided. The plurality of vias 63qb for connecting the second electrode are arranged at intervals in the y direction while being aligned with each other in the x direction. In the present embodiment, the vias 63qb for connecting the second electrode are arranged at positions aligned with the vias 63qa for connecting the second element in the y direction. That is, both the vias 63qb for connecting the second electrode and the vias 63qa for connecting the second element are arranged at positions shifted from the via conductor 62 for ground in the y direction. Also, both the vias 63qb for connecting the second electrode and the vias 63qa for connecting the second element are arranged at positions shifted from the vias 63pa for connecting the first element and the vias 63pb for connecting the first electrode in the y direction. Note that the arrangement position of the vias 63qb for connecting the second electrode in the y direction can be arbitrarily changed. In one example, the vias 63qb for connecting the second electrode may be arranged at positions shifted from the vias 63qa for connecting the second element in the y direction. At least one of the plurality of vias 63qb for connecting the second electrode may be arranged at a position aligned with the via conductor 62 for ground in the y direction.

[0079] Viewed from the z direction, the via 63qb for connecting the second electrode is disposed at a position overlapping with the output electrode 73. More specifically, viewed from the z direction, the via 63qb for connecting the second electrode is disposed at a position overlapping with the end portion closer to the resin side surface 56 among both end portions in the x direction of the output electrode 73. Also, viewed from the z direction, the via 63qb for connecting the second electrode is disposed at a position overlapping with the wiring 63qc for connecting the second via. In the present embodiment, the via 63qb for connecting the second electrode is disposed at a position overlapping with the end portion closer to the resin side surface 55 (the end portion closer to the output electrode 73) among both end portions in the x direction of the wiring 63qc for connecting the second via. Further, the end portion closer to the resin side surface 55 among both end portions in the x direction of the wiring 63qc for connecting the second via is a portion overlapping with the output electrode 73 when viewed from the z direction. In other words, viewed from the z direction, the via 63qb for connecting the second electrode is disposed at a position overlapping with the protruding portion of the wiring 63qc for connecting the second via. Thus, viewed from the z direction, the via 63qb for connecting the second electrode is disposed at a position not overlapping with the second switching element 30B. Viewed from the z direction, the via 63qb for connecting the second electrode is disposed between the first switching element 30A and the second switching element 30B in the x direction.

[0080] As shown in FIG. 5, the via 63qb for connecting the second electrode is disposed at a position shifted from the via 63qa for connecting the second element in the z direction. The via 63qb for connecting the second electrode is provided so as to penetrate the resin layer 50B on the surface side in the z direction. More specifically, the resin layer 50 is provided with a through hole 58f that penetrates the resin layer 50B on the surface side in the z direction. Through the through hole 58f, the wiring 63qc for connecting the second via is exposed in the z direction. The through hole 58f exposes, in the z direction, the end portion (the end portion closer to the output electrode 73) of the both end portions in the x direction of the wiring 63qc for connecting the second via, which is closer to the resin side surface 55. The via 63qb for connecting the second electrode is provided so as to fill the through hole 58f. For this reason, the via 63qb for connecting the second electrode extends along the z direction and is in contact with the wiring 63qc for connecting the second via. Further, the via 63qb for connecting the second electrode is exposed from the resin layer 50 in the z direction. In other words, the via 63qb for connecting the second electrode is connected to the end portion (the end portion closer to the resin side surface 55) of the both end portions in the x direction of the wiring 63qc for connecting the second via, which is closer to the output electrode 73. Then, the via 63qb for connecting the second electrode is exposed from a position overlapping with the end portion (the end portion closer to the resin side surface 55) of the both end portions in the x direction of the wiring 63qc for connecting the second via, which is closer to the output electrode 73, when viewed from the z direction in the element sealing layer 50A. The portion of the via 63qb for connecting the second electrode exposed from the resin layer 50 is covered by the output electrode 73. For this reason, the via 63qb for connecting the second electrode is in contact with the output electrode 73. That is, the via 63qb for connecting the second electrode connects both the wiring 63qc for connecting the second via and the output electrode 73.

[0081] As shown in FIG. 3, the shape of the via 63qb for connecting the second electrode when viewed from the z direction is circular. In the present embodiment, the diameter of the via 63qb for connecting the second electrode is equal to the diameter of the via 63qa for connecting the second element, for example, 100 μm or more and 200 μm or less. Note that the shape of the via 63qb for connecting the second electrode when viewed from the z direction is not limited to a circle and can be arbitrarily changed. For example, the shape of the via 63qb for connecting the second electrode when viewed from the z direction may be a polygon such as a quadrangle or an ellipse.

[0082] In this embodiment, the length of the via 63qb for connecting the second electrode in the z direction is longer than the length of the via 63qa for connecting the second element in the z direction. Note that the length of the via 63qb for connecting the second electrode in the z direction can be arbitrarily changed. In one example, the length of the via 63qb for connecting the second electrode in the z direction may be equal to the length of the via 63qa for connecting the second element in the z direction. In this case, the thickness of the portion of the element sealing layer 50A that covers the second element main surface 31B of the second switching element 30B is equal to the thickness of the surface-side resin layer 50B.

[0083] In this embodiment, the number, shape, and size of the vias 63qb for connecting the second electrode are the same as the number, shape, and size of the vias 63qb for connecting the second electrode. In this way, the first output via conductor 63P and the second output via conductor 63Q are provided so that the variation between the first inductance in the conductive path between the second drive pad electrode 31AB of the first switching element 30A and the output electrode 73 and the second inductance in the conductive path between the first drive pad electrode 31BA of the second switching element 30B and the output electrode 73 is reduced. In terms of design, the first output via conductor 63P and the second output via conductor 63Q are provided so that the first inductance and the second inductance are equal to each other.

[0084] As shown in FIGS. 3 and 7, the element control via conductor 64 is a conductor that electrically connects the driver 40 and the control pad electrodes 31AC and 31BC of the switching elements 30A and 30B individually. The element control via conductor 64 is embedded in the resin layer 50. That is, the element control via conductor 64 is not exposed from the resin layer 50.

[0085] The element control via conductor 64 has a first control via conductor 64P that connects the driver pad electrode 43 of the driver 40 and the control pad electrode 31AC of the first switching element 30A, and a second control via conductor 64Q that connects another driver pad electrode 43 of the driver 40 and the control pad electrode 31BC of the second switching element 30B.

[0086] As shown in FIG. 7, the first control via conductor 64P includes a first element-side control via 64pa connected to the control pad electrode 31AC, a first driver-side control via 64pb connected to the driver pad electrode 43 of the driver 40, and a first control connection wiring 64pc connecting the first element-side control via 64pa and the first driver-side control via 64pb.

[0087] As shown in FIG. 3, when viewed from the z direction, the first element-side control via 64pa is disposed at a position overlapping the control pad electrode 31AC. As shown in FIG. 7, the first element-side control via 64pa is provided so as to penetrate in the z direction through a portion between the first element main surface 31A of the first switching element 30A in the element sealing layer 50A and the interface 57 between the element sealing layer 50A and the surface-side resin layer 50B. In other words, the first element-side control via 64pa penetrates through a portion of the element sealing layer 50A that covers the first element main surface 31A. More specifically, a through hole 58g is provided in the element sealing layer 50A so as to penetrate through a portion in the z direction between the first element main surface 31A of the first switching element 30A and the interface 57 in the element sealing layer 50A. The control pad electrode 31AC is exposed in the z direction from the element sealing layer 50A through the through hole 58g. The first element-side control via 64pa is provided so as to fill the through hole 58g. Therefore, the first element-side control via 64pa extends along the z direction and is in contact with the control pad electrode 31AC. The length of the first element-side control via 64pa in the z direction in the present embodiment is shorter than the length of the first switching element 30A in the z direction. The length of the first element-side control via 64pa in the z direction is less than about 1 mm. In one example, the length of the first element-side control via 64pa in the z direction is about several hundred μm. In the present embodiment, the length of the first element-side control via 64pa in the z direction is equal to the length of the first element connection via 63pa in the z direction.

[0088] As shown in FIG. 3, the shape of the first element side control via 64pa viewed from the z direction is circular. The diameter of the first element side control via 64pa is, for example, 100 μm or more and 200 μm or less. In the present embodiment, the diameter of the first element side control via 64pa is smaller than the diameter of the vias through which the drive currents of the switching elements 30A and 30B such as the power supply via conductor 61 and the ground via conductor 62 flow. Note that the shape of the first element side control via 64pa viewed from the z direction is not limited to a circle and can be arbitrarily changed. For example, the shape of the first element side control via 64pa viewed from the z direction may be a polygon such as a quadrangle or an ellipse. Also, the diameter of the first element side control via 64pa can be arbitrarily changed. In one example, the diameter of the first element side control via 64pa may be equal to the diameter of the vias through which the drive currents of the switching elements 30A and 30B such as the power supply via conductor 61 flow.

[0089] As shown in FIG. 3, when viewed from the z direction, the first driver side control via 64pb is disposed at a position overlapping a predetermined driver pad electrode 43 of the driver 40. In the present embodiment, the predetermined driver pad electrode 43 is provided at a position facing the control pad electrode 31AC in the driver main surface 41 of the driver 40 in the y direction. Specifically, the predetermined driver pad electrode 43 is provided at a corner of the driver main surface 41 that is close to the resin side surface 55 and the resin side surface 54 among the four corners when viewed from the z direction. In the present embodiment, the first driver side control via 64pb is disposed at a position aligned with the first element side control via 64pa in the x direction.

[0090] As shown in FIG. 7, the first driver-side control via 64pb penetrates in the z direction through a portion of the element encapsulation layer 50A that covers the driver main surface 41. More specifically, the element encapsulation layer 50A is provided with a through hole 58h that penetrates in the z direction through a portion of the element encapsulation layer 50A that covers the driver main surface 41. Through the through hole 58h, a predetermined driver pad electrode 43 is exposed in the z direction from the element encapsulation layer 50A. The first driver-side control via 64pb is provided so as to fill the through hole 58h. For this reason, the first driver-side control via 64pb extends along the z direction and is in contact with a predetermined driver pad electrode 43. The length of the first driver-side control via 64pb in the z direction in the present embodiment is shorter than the length of the driver 40 in the z direction. The length of the first driver-side control via 64pb in the z direction is less than about 1 mm. In one example, the length of the first driver-side control via 64pb in the z direction is about several hundred μm.

[0091] As shown in FIG. 3, the shape of the first driver-side control via 64pb viewed from the z direction is circular. The diameter of the first driver-side control via 64pb is equal to the diameter of the first element-side control via 64pa and is, for example, 100 μm or more and 200 μm or less. Note that the shape of the first driver-side control via 64pb viewed from the z direction is not limited to circular and can be arbitrarily changed. For example, the shape of the first driver-side control via 64pb viewed from the z direction may be a polygon such as a quadrangle or an ellipse.

[0092] As shown in FIG. 7, the first control connection wiring 64pc is provided on the element encapsulation layer 50A. In other words, the first control connection wiring 64pc is provided at the interface 57 between the element encapsulation layer 50A and the surface-side resin layer 50B. As shown in FIG. 3, the first control connection wiring 64pc is in a strip shape extending along the y direction.

[0093] As shown in FIGS. 3 and 7, among both end portions in the y direction of the first control connection wiring 64pc, the end portion on the first switching element 30A side covers the portion exposed in the z direction from the element sealing layer 50A among the first element side control vias 64pa. For this reason, the first control connection wiring 64pc is in contact with the first element side control via 64pa. Thus, when viewed from the z direction, the first element side control via 64pa is disposed at a position overlapping both the control pad electrode 31AC and the first control connection wiring 64pc. Further, among both end portions in the y direction of the first control connection wiring 64pc, the end portion on the driver 40 side covers the portion exposed in the z direction from the element sealing layer 50A among the first driver side control vias 64pb. For this reason, the first control connection wiring 64pc is in contact with the first driver side control via 64pb. Thus, when viewed from the z direction, the first driver side control via 64pb is disposed at a position overlapping both a predetermined driver pad electrode 43 and the first control connection wiring 64pc.

[0094] As shown in FIG. 3, the second control via conductor 64Q includes a second element side control via 64qa connected to the control pad electrode 31BC of the second switching element 30B, a second driver side control via 64qb connected to another driver pad electrode 43 of the driver 40, and a second control connection wiring 64qc connecting the second element side control via 64qa and the second driver side control via 64qb.

[0095] As shown in FIG. 3, when viewed from the z direction, the second element side control via 64qa is disposed at a position overlapping the control pad electrode 31BC. The second element side control via 64qa penetrates in the z direction through a portion of the element sealing layer 50A that covers the second element main surface 31B. More specifically, the element sealing layer 50A is provided with a through hole 58i that penetrates in the z direction through a portion of the element sealing layer 50A that covers the second element main surface 31B. The control pad electrode 31BC is exposed in the z direction from the element sealing layer 50A through the through hole 58i. The second element side control via 64qa is provided so as to fill the through hole 58i. Therefore, the second element side control via 64qa extends along the z direction and is in contact with the control pad electrode 31BC. The length of the second element side control via 64qa in the z direction in the present embodiment is shorter than the length of the second switching element 30B in the z direction. The length of the second element side control via 64qa in the z direction is equal to the length of the first element side control via 64pa in the z direction and is less than about 1 mm. In one example, the length of the second element side control via 64qa in the z direction is about several hundred μm. Here, if the difference between the length of the second element side control via 64qa in the z direction and the length of the first element side control via 64pa in the z direction is, for example, 10% or less of the length of the second element side control via 64qa in the z direction, it can be said that the length of the second element side control via 64qa in the z direction is equal to the length of the first element side control via 64pa in the z direction.

[0096] As shown in FIG. 3, the shape of the second element side control via 64qa when viewed from the z direction is circular. The diameter of the second element side control via 64qa is equal to the diameter of the first element side control via 64pa and is, for example, 100 μm or more and 200 μm or less. Note that the shape of the second element side control via 64qa when viewed from the z direction is not limited to circular and can be arbitrarily changed. For example, the shape of the second element side control via 64qa when viewed from the z direction may be a polygon such as a quadrangle or an ellipse.

[0097] As shown in FIG. 3, when viewed from the z-direction, the second driver-side control via 64qb is disposed at a position overlapping with another driver pad electrode 43 of the driver 40. In the present embodiment, the other driver pad electrode 43 is provided near the control pad electrode 31BC on the driver main surface 41 of the driver 40. Specifically, the other driver pad electrode 43 is provided at a corner near the resin side surface 56 and the resin side surface 54 among the four corners of the driver main surface 41 when viewed from the z-direction. In the present embodiment, the second driver-side control via 64qb is disposed at a position shifted in the x-direction with respect to the second element-side control via 64qa. The second driver-side control via 64qb is disposed closer to the support side surface 26 than the second element-side control via 64qa in the x-direction.

[0098] The second driver-side control via 64qb penetrates in the z-direction a portion of the element encapsulation layer 50A that covers the driver main surface 41. More specifically, the element encapsulation layer 50A is provided with a through-hole 58j that penetrates in the z-direction a portion of the element encapsulation layer 50A that covers the driver main surface 41. Through the through-hole 58j, another driver pad electrode 43 is exposed in the z-direction from the element encapsulation layer 50A. The second driver-side control via 64qb is provided so as to fill the through-hole 58j. For this reason, the second driver-side control via 64qb extends along the z-direction and is in contact with another driver pad electrode 43. The length of the second driver-side control via 64qb in the z-direction in the present embodiment is shorter than the length of the driver 40 in the z-direction. The length of the second driver-side control via 64qb in the z-direction is equal to the length of the first driver-side control via 64pb in the z-direction and is less than about 1 mm. In one example, the length of the second driver-side control via 64qb in the z-direction is about several hundred μm. Here, if the difference between the length of the second driver-side control via 64qb in the z-direction and the length of the first driver-side control via 64pb in the z-direction is, for example, 10% or less of the length of the second driver-side control via 64qb in the z-direction, it can be said that the length of the second driver-side control via 64qb in the z-direction is equal to the length of the first driver-side control via 64pb in the z-direction.

[0099] As shown in FIG. 3, the shape of the first driver-side control via 64pb viewed from the z direction is circular. The diameter of the first driver-side control via 64pb is equal to the diameter of the first element-side control via 64pa, for example, 100 μm or more and 200 μm or less. Note that the shape of the first driver-side control via 64pb viewed from the z direction is not limited to a circle and can be arbitrarily changed. For example, the shape of the first driver-side control via 64pb viewed from the z direction may be a polygon such as a quadrilateral or an ellipse.

[0100] The first control connection wiring 64pc is provided on the element encapsulation layer 50A. In other words, the first control connection wiring 64pc is provided at the interface 57 between the element encapsulation layer 50A and the surface-side resin layer 50B. As shown in FIG. 3, the first control connection wiring 64pc is formed in a strip shape and has a portion that obliquely extends toward the resin side surface 56 as it goes from the driver 40 toward the second switching element 30B in the y direction.

[0101] As shown in FIG. 3, the end portion on the second switching element 30B side of both end portions in the y direction of the second control connection wiring 64qc covers the portion of the second element-side control via 64qa that is exposed in the z direction from the element encapsulation layer 50A. For this reason, the second control connection wiring 64qc is in contact with the second element-side control via 64qa. In this way, when viewed from the z direction, the second element-side control via 64qa is disposed at a position overlapping both the control pad electrode 31BC and the second control connection wiring 64qc. Also, the end portion on the driver 40 side of both end portions in the y direction of the second control connection wiring 64qc covers the portion of the second driver-side control via 64qb that is exposed in the z direction from the element encapsulation layer 50A. For this reason, the second control connection wiring 64qc is in contact with the second driver-side control via 64qb. In this way, when viewed from the z direction, the second driver-side control via 64qb is disposed at a position overlapping both another driver pad electrode 43 and the second control connection wiring 64qc.

[0102] As shown in FIGS. 3 and 8, a plurality of driver vias 65 are provided. The driver vias 65 are conductors that individually and electrically connect a plurality of driver pad electrodes 43 of the driver 40 and a plurality of driver connection electrodes 74.

[0103] As shown in FIG. 8, the driver vias 65 are provided so as to penetrate the resin layer 50 in the z direction. In other words, the driver vias 65 are provided so as to penetrate both the element sealing layer 50A and the surface-side resin layer 50B in the z direction. As shown in FIG. 8, the driver vias 65 are crank-shaped. The driver vias 65 include a driver connection via 65a electrically connected to the driver 40, a driver electrode connection via 65b electrically connected to the driver connection electrode 74, and a via connection wiring 65c connecting the driver connection via 65a and the driver electrode connection via 65b.

[0104] As shown in FIG. 3, in the present embodiment, when viewed from the z direction, the driver connection via 65a is disposed at a position overlapping the driver pad electrode 43 of the driver 40.

[0105] As shown in FIG. 8, the driver connection via 65a penetrates in the z direction through the portion of the element sealing layer 50A that covers the driver main surface 41. More specifically, the element sealing layer 50A is provided with a through hole 58k that penetrates in the z direction through the portion of the element sealing layer 50A that covers the driver main surface 41. The driver pad electrode 43 is exposed in the z direction through the through hole 58k. The driver connection via 65a is provided so as to fill the through hole 58k. Therefore, the driver connection via 65a extends along the z direction and is in contact with the driver pad electrode 43. The length of the driver connection via 65a in the z direction in the present embodiment is shorter than the length of the driver 40 in the z direction. The length of the driver connection via 65a in the z direction is equal to the length of the first driver-side control via 64pb in the z direction and is less than about 1 mm. In one example, the length of the driver connection via 65a in the z direction is about several hundred micrometers. Here, if the difference between the length of the driver connection via 65a in the z direction and the length of the first driver-side control via 64pb in the z direction is, for example, 10% or less of the length of the driver connection via 65a in the z direction, it can be said that the length of the driver connection via 65a in the z direction is equal to the length of the first driver-side control via 64pb in the z direction.

[0106] As shown in FIG. 3, the shape of the driver connection via 65a viewed from the z direction is circular. The diameter of the driver connection via 65a is, for example, 100 μm or more and 200 μm or less. Note that the shape of the driver connection via 65a viewed from the z direction is not limited to circular and can be arbitrarily changed. For example, the shape of the driver connection via 65a viewed from the z direction may be a polygon such as a quadrilateral or an ellipse.

[0107] As shown in FIG. 8, the via connection wiring 65c is provided on the element encapsulation layer 50A. In other words, the via connection wiring 65c is provided at the interface 57 between the element encapsulation layer 50A and the surface-side resin layer 50B. As shown in FIG. 3, the shape of the via connection wiring 65c viewed from the z direction is strip-shaped. Viewed from the z direction, the via connection wiring 65c covers the driver connection via 65a. In other words, the driver connection via 65a is arranged at a position overlapping with the via connection wiring 65c when viewed from the z direction. More specifically, the driver connection via 65a is arranged at a position overlapping both the driver pad electrode 43 and the via connection wiring 65c when viewed from the z direction.

[0108] As shown in FIG. 3, when viewed from the z direction, the driver electrode connection via 65b is arranged at a position overlapping with the driver connection electrode 74. Also, when viewed from the z direction, the driver electrode connection via 65b is arranged at a position overlapping with the via connection wiring 65c.

[0109] As shown in FIG. 8, the driver electrode connection via 65b is provided so as to penetrate the surface-side resin layer 50B in the z direction. More specifically, the resin layer 50 is provided with a through hole 58l that penetrates the surface-side resin layer 50B in the z direction. The via connection wiring 65c is exposed in the z direction through the through hole 58l. The driver electrode connection via 65b is provided so as to fill the through hole 58l. For this reason, the driver electrode connection via 65b extends along the z direction and is in contact with the via connection wiring 65c. Also, the driver electrode connection via 65b is exposed from the resin layer 50 in the z direction. The portion of the driver electrode connection via 65b that is exposed from the resin layer 50 is covered by the driver connection electrode 74. For this reason, the driver electrode connection via 65b is in contact with the driver connection electrode 74.

[0110] In this embodiment, the length of the via 65b for driver electrode connection in the z direction is longer than the length of the via 65a for driver connection in the z direction. The length of the via 65b for driver electrode connection in the z direction is equal to the length of each of the vias 63pb and 63qb for electrode connection in the z direction. Here, if the difference between the length of the via 65b for driver electrode connection in the z direction and the length of each of the vias 63pb and 63qb for electrode connection in the z direction is, for example, 10% or less of the length of the via 65b for driver electrode connection in the z direction, it can be said that the length of the via 65b for driver electrode connection in the z direction is equal to the length of each of the vias 63pb and 63qb for electrode connection in the z direction. Note that the length of the via 65b for driver electrode connection in the z direction can be arbitrarily changed. In one example, the length of the via 65b for driver electrode connection in the z direction may be equal to the length of the via 65a for driver connection in the z direction.

[0111] As shown in FIG. 3, the shape of the via 65b for driver electrode connection viewed from the z direction is circular. The diameter of the via 65b for driver electrode connection is equal to the diameter of the via 65a for driver connection, for example, 100 μm or more and 200 μm or less. Note that the shape of the via 65b for driver electrode connection viewed from the z direction is not limited to circular and can be arbitrarily changed. For example, the shape of the via 65b for driver electrode connection viewed from the z direction may be a polygon such as a quadrilateral or an ellipse.

[0112] Next, with reference to FIG. 9, an example of the circuit configuration of the semiconductor device 10 will be described.

[0113] The plus terminal of a drive power source DV, which is a DC power source, is electrically connected to the first drive pad electrode 31AA (drain electrode) of the first switching element 30A. The second drive pad electrode 31AB (source electrode) of the first switching element 30A is electrically connected to the first drive pad electrode 31BA (drain electrode) of the second switching element 30B. The second drive pad electrode 31BB (source electrode) of the second switching element 30B is connected to the ground. In this way, the first switching element 30A and the second switching element 30B are connected in series.

[0114] A capacitor C1 is connected to the series combination of the first switching element 30A and the second switching element 30B. The capacitor C1 has a function of removing noise from the voltage supplied from the drive power source DV to the first drive pad electrode 31AA of the first switching element 30A.

[0115] A node N between the second drive pad electrode 31AB of the first switching element 30A and the first drive pad electrode 31BA of the second switching element 30B is electrically connected to a load L to which the semiconductor device 10 supplies power. The load L is provided outside the semiconductor device 10. An example of the load L is a motor.

[0116] The driver 40 is electrically connected to the control pad electrode 31AC (gate electrode) of the first switching element 30A and the control pad electrode 31BC (gate electrode) of the second switching element 30B. Based on a signal from a signal generation circuit (not shown) provided outside the semiconductor device 10, the driver 40 generates a gate voltage for controlling the on / off operation of each of the switching elements 30A and 30B and supplies it to each of the control pad electrodes 31AC and 31BC.

[0117] The plurality of driver pad electrodes 43 of the driver 40 include a first signal input terminal HIN, a second signal input terminal LIN, a control-side power supply electrode VCC, a control-side ground electrode GND, a bootstrap terminal VB, a first signal output electrode HOH, a second signal output electrode HOL, an output-side power supply electrode VS, a third signal output electrode LOH, a fourth signal output electrode LOL, and an output-side ground electrode PGND. Note that the plurality of driver pad electrodes 43 may include terminals other than the above-described terminals.

[0118] The first signal input terminal HIN is a terminal to which the high-potential side signal from the signal generation circuit is input. The second signal input terminal LIN is a terminal to which the low-potential side signal from the signal generation circuit is input. The driver 40 generates a gate voltage to be output to each of the control pad electrodes 31AC and 31BC of each of the switching elements 30A and 30B based on the high-potential side signal and the low-potential side signal input from the signal generation circuit via each of the signal input terminals HIN and LIN.

[0119] The control-side power supply electrode VCC is a terminal to which the plus terminal of a control power supply CV, which is a DC power supply provided outside the semiconductor device 10, is electrically connected. The minus terminal of the control power supply CV is connected to the ground. A capacitor C2 is connected in parallel with the control power supply CV between the control-side power supply electrode VCC and the control power supply CV. The capacitor C2 has a function of removing noise of the voltage supplied from the control power supply CV to the control-side power supply electrode VCC. The first terminal of the capacitor C2 is connected to the plus terminal of the control-side power supply electrode VCC, and the second terminal of the capacitor C2 is connected to the ground. The control-side ground electrode GND is connected to the ground. More specifically, the control-side ground electrode GND is electrically connected to the second terminal of the capacitor C2. The driver 40 is driven based on the voltage (for example, 5V) of the control power supply CV.

[0120] The output-side power supply electrode VS is a terminal that serves as the power supply for the load L. The output-side power supply electrode VS is connected to a node N between the second drive pad electrode 31AB of the first switching element 30A and the first drive pad electrode 31BA of the second switching element 30B.

[0121] Each of the first signal output electrode HOH and the second signal output electrode HOL is a terminal that supplies the gate voltage generated in the driver 40 to the control pad electrode 31AC of the first switching element 30A. Each of the signal output electrodes HOH and HOL is electrically connected to the control pad electrode 31AC. A current limiting resistor R1 is provided between the first signal output electrode HOH and the control pad electrode 31AC.

[0122] Each of the third signal output electrode LOH and the fourth signal output electrode LOL is a terminal that supplies the gate voltage generated in the driver 40 to the control pad electrode 31BC of the second switching element 30B. Each of the signal output electrodes LOH and LOL is electrically connected to the control pad electrode 31BC. A current limiting resistor R2 is provided between the third signal output electrode LOH and the control pad electrode 31BC.

[0123] The bootstrap terminal VB is a terminal to which a bootstrap capacitor BC that constitutes a bootstrap circuit (not shown) for generating the gate voltage output from the first signal output electrode HOH and the second signal output electrode HOL to a high potential is connected. A bootstrap diode that constitutes the bootstrap circuit is provided in the driver 40. The first terminal of the bootstrap capacitor BC is connected to the bootstrap terminal VB, and the second terminal of the bootstrap capacitor BC is connected to a node N between the second drive pad electrode 31AB of the first switching element 30A and the first drive pad electrode 31BA of the second switching element 30B.

[0124] The output side ground electrode PGND is electrically connected to the second drive pad electrode 31BB (source electrode) of the second switching element 30B. That is, the output side ground electrode PGND is connected to the ground.

[0125] According to this configuration, when the high potential side signal and the low potential side signal from the signal generation circuit are input to the driver 40, the driver 40 generates a gate voltage based on these signals and supplies it to each of the control pad electrodes 31AC and 31BC of each of the switching elements 30A and 30B. Each of the switching elements 30A and 30B operates complementarily on and off based on the gate voltage.

[0126] (Method of manufacturing a semiconductor device) With reference to FIGS. 10 to 24, an example of a method of manufacturing the semiconductor device 10 of the present embodiment will be described.

[0127] The manufacturing method of the semiconductor device 10 mainly includes a wiring layer formation step, a mounting step, an element encapsulation layer formation step, a first via conductor formation step, a surface-side resin layer formation step, a second via conductor formation step, an external electrode formation step, and a dicing step.

[0128] In the wiring layer formation step, as shown in FIG. 10, first, a support substrate 800 made of, for example, Si (silicon) is prepared. The support substrate 800 has a substrate main surface 801 facing one side in the z direction. Next, a base material 820 is formed on the substrate main surface 801. The base material 820 is a member constituting the support layer 20 of the semiconductor device 10 and is made of, for example, an epoxy resin. The base material 820 is formed, for example, by transfer molding or compression molding. The base material 820 has a base material main surface 821 facing the same side as the substrate main surface 801 in the z direction.

[0129] In the mounting step, each of the switching elements 30A and 30B shown in FIG. 11 and the driver 40 shown in FIG. 12 is mounted on the base material 820.

[0130] As shown in FIG. 11, each of the switching elements 30A and 30B is mounted on the base material main surface 821 of the base material 820. More specifically, a liquid bonding material SD is applied to the positions on the base material main surface 821 where each of the switching elements 30A and 30B is to be mounted. Next, each of the switching elements 30A and 30B is mounted on the bonding material SD, for example, by die bonding. Subsequently, the bonding material SD is cured, whereby the bonding material SD and each of the switching elements 30A and 30B are bonded. When solder or an Ag paste is used as the bonding material SD, each of the switching elements 30A and 30B may be bonded to the base material main surface 821 by a reflow process and a cooling process.

[0131] As shown in FIG. 12, the driver 40 is mounted on the base material main surface 821. The method of mounting the driver 40 on the base material main surface 821 is the same as the method of mounting each of the switching elements 30A and 30B on the base material main surface 821.

[0132] The element encapsulation layer forming step includes a resin layer forming step shown in FIG. 13 and a grinding step shown in FIG. 14.

[0133] As shown in FIG. 13, in the resin layer forming step, an element encapsulation layer 850A for encapsulating each of the switching elements 30A, 30B and the driver 40 is formed. The element encapsulation layer 850A is a layer that constitutes the element encapsulation layer 50A of the semiconductor device 10 and is made of, for example, a black epoxy resin. The thickness of the element encapsulation layer 850A is thicker than the thickness of the element encapsulation layer 50A. The element encapsulation layer 850A is formed by, for example, transfer molding or compression molding.

[0134] As shown in FIG. 14, in the grinding step, the element encapsulation layer 850A is removed in the z direction. Thereby, the thickness of the element encapsulation layer 850A is reduced. More specifically, the portion of the element encapsulation layer 850A on the side opposite to the base material 820 in the z direction is removed by mechanical grinding. Thereby, the distance D between the element main surfaces 31A, 31B of the switching elements 30A, 30B and the driver main surface 41 of the driver 40 and the element encapsulation layer 850A in the z direction is reduced. The element encapsulation layer 850A is removed until the distance D becomes about several hundred μm, for example. Thereby, the thickness of the element encapsulation layer 850A becomes equal to the thickness of the element encapsulation layer 50A. In this way, the thickness of the portion of the element encapsulation layer 850A that covers each of the element main surfaces 31A, 31B of the switching elements 30A, 30B and the driver main surface 41 of the driver 40 is reduced.

[0135] The first via conductor forming step includes a through-hole forming step and a first via forming step shown in FIGS. 15 and 16, and a wiring forming step shown in FIGS. 17 and 18.

[0136] As shown in FIGS. 15 and 16, in the through-hole forming step, through-holes 58c, 58e, 58g to 58k are formed in the element sealing layer 850A by a hole-opening process such as laser processing. In FIG. 15, a through-hole 58c that exposes the second drive pad electrode 31AB of the first switching element 30A in the z direction and a through-hole 58e that exposes the first drive pad electrode 31BA of the second switching element 30B in the z direction are formed in the element sealing layer 850A. The through-holes 58c and 58e penetrate in the z direction through a portion of the element sealing layer 850A that covers the element main surfaces 31A and 31B of the respective switching elements 30A and 30B. As a result, the second drive pad electrode 31AB of the first switching element 30A and the first drive pad electrode 31BA of the second switching element 30B are each exposed in the z direction.

[0137] As shown in FIG. 16, the through-holes 58g to 58k are also formed in the element sealing layer 850A. Each of the through-holes 58g and 58i penetrates in the z direction through a portion of the element sealing layer 850A that covers the element main surfaces 31A and 31B of the respective switching elements 30A and 30B, similar to the through-holes 58c and 58e. As a result, the control pad electrodes 31AC and 31BC of the respective switching elements 30A and 30B are exposed in the z direction. Further, the through-holes 58h, 58j, and the plurality of through-holes 58k penetrate in the z direction through a portion of the element sealing layer 850A that covers the driver main surface 41 of the driver 40. As a result, each driver pad electrode 43 of the driver 40 is exposed in the z direction.

[0138] As shown in FIGS. 15 and 16, in the via formation process, metal vias are embedded in each through hole 58c, 58e, 58g to 58k. More specifically, as shown in FIG. 15, the first element connection via 63pa is formed to be embedded in the through hole 58c, and the second element connection via 63qa is formed to be embedded in the through hole 58e. More specifically, a seed layer is formed on the element sealing layer 850A. The seed layer is formed on the element sealing layer 850A by electroless plating. In this case, the seed layer is also formed on the inner surfaces constituting the through holes 58c and 58e. The seed layer is made of, for example, Ti. Next, a plating layer is formed on the seed layer. The plating layer is formed by electrolytic plating using the seed layer as a conductive path after lithographic patterning is performed on the seed layer. The plating layer is made of Cu. The plating layer is formed so as to fill the through holes 58c and 58e. Next, the portion of the seed layer that is not covered by the plating layer is removed. In the present embodiment, the seed layer on the portions other than the inner surfaces constituting the through holes 58c and 58e is removed.

[0139] Also, as shown in FIG. 16, the first element side control via 64pa, which is a via embedded in the through hole 58g, the second element side control via 64qa, which is a via embedded in the through hole 58i, the first driver side control via 64pb, which is a via embedded in the through hole 58h, the second driver side control via 64qb, which is a via embedded in the through hole 58j, and the driver connection via 65a, which is a via embedded in the through hole 58k, are formed in the same manner as the respective element connection vias 63pa and 63qa. In the present embodiment, the respective element connection vias 63pa and 63qa, the respective element side control vias 64pa and 64qa, the respective driver side control vias 64pb and 64qb, and the plurality of driver connection vias 65a are formed simultaneously.

[0140] As shown in FIGS. 17 and 18, in the wiring formation step, the first via connection wiring 63pc, the second via connection wiring 63qc, the first control connection wiring 64pc, the second control connection wiring 64qc, and a plurality of via connection wirings 65c are formed. These wirings 63pc, 63qc, 64pc, 64qc, 65c are formed in the same manner as the respective element connection vias 63pa, 63qa in the via formation step. Thus, since each of the first via connection wiring 63pc, the second via connection wiring 63qc, the first control connection wiring 64pc, the second control connection wiring 64qc, and the plurality of via connection wirings 65c is formed on the element sealing layer 850A, these wirings are formed simultaneously in the wiring formation step. Through the wiring formation step, the first control via conductor 64P and the second control via conductor 64Q are formed.

[0141] Note that, in this embodiment, the via formation step and the wiring formation step are separate steps, but it is not limited thereto. For example, the via formation step and the wiring formation step may be the same step. That is, the respective element connection vias 63pa, 63qa, the respective element-side control vias 64pa, 64qa, the respective driver-side control vias 64pb, 64qb, and the plurality of driver connection vias 65a, and the first via connection wiring 63pc, the second via connection wiring 63qc, the first control connection wiring 64pc, the second control connection wiring 64qc, and the plurality of via connection wirings 65c may be formed simultaneously.

[0142] As shown in FIG. 19, in the surface-side resin layer forming step, a surface-side resin layer 850B is formed on the element sealing layer 850A. The surface-side resin layer 850B is a layer that constitutes the surface-side resin layer 50B of the semiconductor device 10 and is made of, for example, a black epoxy resin. In the present embodiment, the surface-side resin layer 850B is made of the same material as the element sealing layer 850A. When viewed from the z direction, the surface-side resin layer 850B is formed over the entire element sealing layer 850A. The surface-side resin layer 850B is formed so as to seal the first via connection wiring 63pc, the second via connection wiring 63qc, the first control connection wiring 64pc, and the second control connection wiring 64qc. The surface-side resin layer 850B is formed, for example, by transfer molding or compression molding. Thereby, an interface 857 is formed between the element sealing layer 850A and the surface-side resin layer 850B in the z direction. The thickness of the surface-side resin layer 850B is thinner than the thickness of the portion of the element sealing layer 850A other than the portion covering each of the switching elements 30A, 30B and the driver 40. In the present embodiment, the thickness of the surface-side resin layer 850B is thicker than the thickness of the portion of the element sealing layer 850A covering the element main surfaces 31A, 31B of each of the switching elements 30A, 30B.

[0143] Note that the surface-side resin layer forming step may include a grinding step. When the surface-side resin layer forming step includes a grinding step, for example, the surface-side resin layer 850B is removed in the z direction by mechanical grinding. Thereby, the thickness of the surface-side resin layer 850B is made thinner.

[0144] As shown in FIGS. 20 to 22, the second via conductor forming step includes a through hole forming step and a via forming step.

[0145] The through hole forming step includes a step of forming the through holes 58d, 58f shown in FIG. 20, a step of forming the through holes 58a, 58b shown in FIG. 21, and a step of forming the through hole 58l shown in FIG. 22.

[0146] As shown in FIGS. 20 and 22, the through holes 58d, 58f, 58l are formed in the surface-side resin layer 850B by a hole-making process such as laser processing. The through holes 58d, 58f, 58l penetrate only the surface-side resin layer 850B in the z direction. As a result, as shown in FIG. 20, the first via connection wiring 63pc is exposed in the z direction through the through hole 58d, and the second via connection wiring 63qc is exposed in the z direction through the through hole 58f. Further, as shown in FIG. 22, the via connection wiring 65c is exposed in the z direction through the through hole 58l.

[0147] As shown in FIG. 21, the through holes 58a, 58b are formed in the surface-side resin layer 850B and the element sealing layer 850A by a hole-making process such as laser processing. The through holes 58a, 58b penetrate both the surface-side resin layer 850B and the element sealing layer 850A in the z direction. The through holes 58a, 58b penetrate in the z direction the portions of the element sealing layer 850A that cover the respective switching elements 30A, 30B. As a result, the first drive pad electrode 31AA of the first switching element 30A is exposed in the z direction through the through hole 58a, and the second drive pad electrode 31BB of the second switching element 30B is exposed in the z direction through the through hole 58b.

[0148] The via formation process includes a process of forming the first electrode connection via 63pb and the second electrode connection via 63qb shown in FIG. 20, a process of forming the power supply via conductor 61 and the ground via conductor 62 shown in FIG. 21, and a process of forming a plurality of driver electrode connection vias 65b shown in FIG. 22.

[0149] As shown in FIG. 20, the first electrode connection via 63pb is formed to be embedded in the through hole 58d, and the second electrode connection via 63qb is formed to be embedded in the through hole 58f. As shown in FIG. 21, the power supply via conductor 61 is formed to be embedded in the through hole 58a, and the ground via conductor 62 is formed to be embedded in the through hole 58b. As shown in FIG. 22, a plurality of driver electrode connection vias 65b are formed to be embedded in the through hole 58l. The formation methods of the respective electrode connection vias 63pb and 63qb, the power supply via conductor 61, the ground via conductor 62, and the plurality of driver electrode connection vias 65b are the same as the via formation process of the first via conductor formation process. Through the second via conductor formation process, the output via conductor 63 and the driver via conductor 65 are formed.

[0150] As shown in FIGS. 23 and 24, in the exterior electrode formation process, a plating layer is formed on the surface of the surface-side resin layer 850B on the side opposite to the element sealing layer 850A in the z direction. The plating layer includes a Cu layer and a plating layer covering the Cu layer. In forming the Cu layer, first, a seed layer is formed on the surface of the surface-side resin layer 850B on the side opposite to the element sealing layer 850A in the z direction by electroless plating. The seed layer is made of, for example, Cu or Ti. Subsequently, a plating layer is formed by electrolytic plating using the seed layer as a conductive path. This plating layer is made of Cu. The plating layer covering the Cu layer is formed, for example, by electroless plating by depositing, for example, Ni, Pd, and Au in this order. As shown in FIG. 24, the plating layer has a power supply plating layer covering the plurality of power supply via conductors 61, a ground plating layer covering the plurality of ground via conductors 62, an output plating layer covering the plurality of first electrode connection vias 63pb and the plurality of second electrode connection vias 63qb, and a plurality of driver electrode connection plating layers individually covering the plurality of driver electrode connection vias 65b. The power supply plating layer constitutes the power supply electrode 71 as the exterior electrode 70, the ground plating layer constitutes the ground electrode 72 as the exterior electrode 70, the output plating layer constitutes the output electrode 73 as the exterior electrode 70, and the plurality of driver electrode connection plating layers constitute the plurality of driver connection electrodes 74.

[0151] As shown in FIGS. 23 and 24, in the cutting process, first, the support substrate 800 is separated from the base material 820. For example, the support substrate 800 is separated from the base material 820 by mechanical grinding or debonding. Next, a dicing tape (not shown) for supporting the base material 820 is attached. Subsequently, for example, using a dicing blade, the surface-side resin layer 850B, the element encapsulation layer 850A, and the base material 820 are cut in this order along the cutting line CL indicated by the dashed-dotted line in FIGS. 23 and 24. As a result, the element encapsulation layer 50A is formed from the element encapsulation layer 850A, and the surface-side resin layer 50B is formed from the surface-side resin layer 850B. That is, the resin layer 50 is formed. Also, the support layer 20 is formed from the base material 820. Through the above steps, the semiconductor device 10 is manufactured.

[0152] (Operation) With reference to FIGS. 5, 25, and 26, the operation of the semiconductor device 10 of the present embodiment will be described. FIG. 25 is a plan view schematically showing the internal structure of the semiconductor device 10X of the comparative example, and FIG. 26 is a cross-sectional view showing the schematic cross-sectional structure of the semiconductor device 10X of the comparative example.

[0153] As shown in FIG. 25, the semiconductor device 10X of the comparative example includes a support layer 20X, a first switching element 30A, a second switching element 30B, a driver 40, and a resin layer 50X, similar to the semiconductor device 10 of the present embodiment. Each of the switching elements 30A, 30B and the driver 40 is mounted on metal support plates 101 to 103 provided on the support layer 20X. Each of the support plates 101 to 103 is exposed from both the support surface 21X and the support back surface 22X (see FIG. 26) of the support layer 20X.

[0154] Also, the support layer 20X is provided with a power supply wiring 104, a ground wiring 105, an output wiring 106, and a plurality of driver connection wirings 107. These wirings 104 to 107 are also exposed from the support back surface 22X (see FIG. 26) of the support layer 20X, similar to the support plates 101 to 103.

[0155] The support plate 101 that supports the first switching element 30A, the support plate 102 that supports the second switching element 30B, the power supply wiring 104, the ground wiring 105, and the output wiring 106 are arranged at intervals in the x-direction while being aligned with each other in the y-direction. The output wiring 106 is arranged between the support plates 101 and 102 in the x-direction. The output wiring 106 is arranged at the center in the x-direction of the support layer 20X. The power supply wiring 104 is arranged closer to the support side surface 25X than the support plate 101. The ground wiring 105 is arranged closer to the support side surface 26X than the support plate 102. Thus, the support plates 101 and 102 and the respective wirings 104 to 107 are arranged on the same plane.

[0156] The support plate 103 that supports the driver 40 is arranged at the center in the x-direction of the support layer 20X. Also, the support plate 103 is arranged closer to the support side surface 23X than the respective support plates 101 and 102 in the y-direction. In other words, the support plates 101 and 102 are arranged closer to the support side surface 24X than the support plate 103 in the y-direction. A plurality of driver connection wirings 107 are arranged closer to the support side surface 25X than the driver 40 in the x-direction of the support layer 20X. A plurality of driver connection wirings 107 are arranged closer to the support side surface 26X than the driver 40 in the x-direction of the support layer 20X. These driver connection wirings 107 are arranged at the end closer to the support side surface 23X among both ends in the y-direction of the support layer 20X.

[0157] As shown in FIG. 25, the first switching element 30A, the power supply wiring 104, the output wiring 106, and the driver 40 are electrically connected by a wire W. More specifically, each first drive pad electrode 31AA of the first switching element 30A and the power supply wiring 104 are connected by a wire W. Each second drive pad electrode 31AB of the first switching element 30A and the output wiring 106 are connected by a wire W. The control pad electrode 31AC of the first switching element 30A and the driver pad electrode 43 of the driver 40 are connected by a wire W.

[0158] The second switching element 30B is electrically connected to the ground wiring 105, the output wiring 106, and the driver 40 by wires W. More specifically, each first drive pad electrode 31BA of the second switching element 30B is connected to the output wiring 106 by a wire W. The second drive pad electrode 31BB of the second switching element 30B is connected to the ground wiring 105 by a wire W. The control pad electrode 31BC of the second switching element 30B is connected to the driver pad electrode 43 of the driver 40 by a wire W.

[0159] Also, a plurality of driver pad electrodes 43 of the driver 40 and a plurality of driver connection wirings 107 are individually connected by wires W.

[0160] These wires W are formed by wire bonding using a wire bonding apparatus. In one example, the wire bonding apparatus bonds a wire base material to the second drive pad electrode 31AB of the first switching element 30A. Then, the wire bonding apparatus separates the wire base material from the second drive pad electrode 31AB in the z direction and moves it toward the output wiring 106. Then, after bonding the wire base material to the output wiring 106, the wire bonding apparatus cuts the wire base material. Thereby, the wire W is formed. For this reason, as shown in FIG. 26, the wire W connecting the second drive pad electrode 31AB and the output wiring 106 is formed to be curved convex on the side opposite to the support layer 20X in the z direction from the first element main surface 31A. Note that the other wires W also have the same shape.

[0161] Thus, since the wire W is curved convex, the maximum value DW of the distance in the z direction between each element main surface 31A, 31B of each switching element 30A, 30B and the wire W increases. This maximum value DW is larger than the length in the z direction (the thickness of each switching element 30A, 30B) of each switching element 30A, 30B and is, for example, 1 mm or more. For this reason, the length of the wire W connected to each switching element 30A, 30B becomes long.

[0162] The via conductor 60 and the exterior electrode 70 are formed by the same processes and apparatuses as those in the pre-process of the semiconductor manufacturing process (the process of forming semiconductor elements). The wire W is formed by the post-process of the semiconductor manufacturing process (the process of packaging semiconductor elements). Therefore, the wire W has a larger dimensional error compared to the formation of the via conductor 60 and the formation of the exterior electrode 70. More specifically, for example, the positional accuracy when connecting the wire W to the second drive pad electrode 31AB of the first switching element 30A and the positional accuracy when connecting the wire W connected to the second drive pad electrode 31AB to the output wiring 106 are low, so the variation in the length of the wire W becomes large. Also, for example, the positional accuracy when connecting the wire W to the first drive pad electrode 31BA of the second switching element 30B and the positional accuracy when connecting the wire W connected to the first drive pad electrode 31BA to the output wiring 106 are low, so the variation in the length of the wire W becomes large. That is, in design, it is necessary to consider the variation in the length of the wire W, and it is necessary to increase the length of the wire W. Due to such constraints, the length of each wire W is several millimeters (for example, 3 mm or more and 5 mm or less).

[0163] Also, the parasitic inductance of the wire W connecting the control pad electrodes 31AC, 31BC and the driver pad electrode 43 becomes a factor of the surge included in the gate voltage supplied to the control pad electrodes 31AC, 31BC of the switching elements 30A, 30B, and increases as the length of the wire W increases. That is, as the length of the wire W connecting the control pad electrodes 31AC, 31BC and the driver pad electrode 43 increases, there is a possibility that the surge included in the gate voltage increases.

[0164] Also, since the surge is proportional to the rate of change of the gate voltage, it increases as the operation of each of the switching elements 30A and 30B becomes faster. For this reason, if each of the switching elements 30A and 30B is a switching element that does not operate at high frequencies, such as an IGBT (Insulated Gate Bipolar Transistor) or a bipolar transistor, the surge included in the gate voltage does not become excessively large. Therefore, the influence on the operation of each of the switching elements 30A and 30B due to the surge included in the gate voltage is reduced. On the other hand, if each of the switching elements 30A and 30B is a switching element that operates at high frequencies, such as a GaN HEMT or a SiC MOSFET, the surge included in the gate voltage may become excessively large. Therefore, the influence on the operation of each of the switching elements 30A and 30B due to the surge included in the gate voltage becomes large. In one example, in the case of a switching element that operates at high frequencies, such as a GaN HEMT, the inductance of several nH due to the wire W may greatly affect the operation of each of the switching elements 30A and 30B.

[0165] Due to such circumstances, from the viewpoint of reducing the surge included in the gate voltage, it is necessary to reduce the inductance caused by the conductive path between the control pad electrodes 31AC and 31BC and the driver pad electrode 43. In particular, in the case of a switching element that operates at high frequencies, such as a GaN HEMT, the inductance caused by each conductive path between the control pad electrodes 31AC and 31BC and the driver pad electrode 43 of the driver 40 is preferably less than 1 nH.

[0166] In view of such points, in this embodiment, each drive pad electrode 31AA, 31AB, 31BA, 31BB of each switching element 30A, 30B is connected to the exterior electrode 70 by a power supply via conductor 61, a ground via conductor 62, and an output via conductor 63. Each control pad electrode 31AC, 31BC of each switching element 30A, 30B is connected to the driver pad electrode 43 of the driver 40 by an element control via conductor 64. The driver pad electrode 43 of the driver 40 is connected to the exterior electrode 70 by a driver via conductor 65.

[0167] Since these via conductors 61 to 65 are formed by the same process as the pre-process of semiconductor manufacturing, the dimensional accuracy regarding the formation of the via conductors 61 to 65 is high. Specifically, the dimensional accuracy when these via conductors 61 to 65 are joined to each of the pad electrodes 31AA to 31AC, 31BA to 31BC, 43 and the exterior electrode 70 is 1 / 10 or less compared to the dimensional accuracy when the wire W is joined to each of the pad electrodes 31AA to 31AC, 31BA to 31BC, 43 and the exterior electrode 70. Therefore, since the lengths of these via conductors 61 to 65 can be shortened, the inductance caused by the lengths of these via conductors 61 to 65 can be reduced.

[0168] In addition, the cross-sectional area of each element connection via 63pa, 63qa of the power supply via conductor 61, the ground via conductor 62, and the output via conductor 63, each electrode connection via 63pb, 63qb, each element-side control via 64pa, 64qa of the element control via conductor 64, each driver-side control via 64pb, 64qb, the driver connection via 65a of the driver via conductor 65, and the driver electrode connection via 65b, when cut by a plane (xy plane) orthogonal to the z direction, is larger than the cross-sectional area of the wire W cut by a plane orthogonal to its length direction. For this reason, each of the via conductors 61, 62 and each of the vias 63pa, 63qa, 63pb, 63qb, 64pa, 64qa, 65a, 65b has a smaller inductance than the wire W.

[0169] Also, when viewed from the z - direction, the power - supply electrode 71 is arranged so as to overlap with the first switching element 30A, the power - supply via conductor 61 is arranged at a position overlapping both the power - supply electrode 71 and the first drive pad electrode 31AA of the first switching element 30A, and connects the power - supply electrode 71 and the first drive pad electrode 31AA. The ground electrode 72 is arranged so as to overlap with the second switching element 30B, the ground via conductor 62 is arranged at a position overlapping both the ground electrode 72 and the second drive pad electrode 31BB of the second switching element 30B, and connects the ground electrode 72 and the second drive pad electrode 31BB. Thereby, the power - supply via conductor 61 and the ground via conductor 62 are formed in a shape extending in the z - direction, and the length can be further shortened. Therefore, the inductance caused by the length of each via conductor 61, 62 can be reduced.

[0170] In addition, the above - mentioned maximum value DW of the wire W is longer than the lengths of the power - supply via conductor 61 and the ground via conductor 62 in the z - direction. For this reason, the length of the power - supply via conductor 61 in the z - direction is sufficiently shorter than the length of the wire W connecting the first drive pad electrode 31AA and the power - supply wiring 104. The length of the ground via conductor 62 in the z - direction is sufficiently shorter than the length of the wire W connecting the second drive pad electrode 31BB and the ground wiring 105. For this reason, both the inductance caused by the length of the conductive path between the first drive pad electrode 31AA and the power - supply electrode 71 and the inductance caused by the length of the conductive path between the second drive pad electrode 31BB and the ground electrode 72 can be further reduced.

[0171] (Effect) According to the semiconductor device 10 of the present embodiment, the following effects can be obtained.

[0172] (1) The semiconductor device 10 includes a first switching element 30A, a second switching element 30B, a driver 40, a resin layer 50 that encapsulates each of the switching elements 30A, 30B and the driver 40, a power electrode 71 and a ground electrode 72 as external electrodes 70 formed on a resin main surface 51 of the resin layer 50 and at least partially overlapping with each of the switching elements 30A, 30B, and a power via conductor 61 and a ground via conductor 62 as via conductors 60 that penetrate the resin layer 50 in the z direction and electrically connect each of the switching elements 30A, 30B and the external electrodes 70.

[0173] According to this configuration, since the power via conductor 61 and the ground via conductor 62 that penetrate the resin layer 50 in the z direction connect each of the switching elements 30A, 30B and the power electrode 71 and the ground electrode 72 as the external electrodes 70, compared with a configuration in which each of the switching elements 30A, 30B and the power wiring 104 and the ground wiring 105 as the external electrodes are connected by a wire W formed by wire bonding, both the length of the conductive path between the first switching element 30A and the power electrode 71 and the length of the conductive path between the second switching element 30B and the ground electrode 72 can be shortened. Therefore, the inductance caused by the lengths of these conductive paths can be reduced.

[0174] In addition, when viewed from the z direction, since at least a part of the power electrode 71 is disposed at a position overlapping with the first switching element 30A and at least a part of the ground electrode 72 is disposed at a position overlapping with the second switching element 30B, the size of the semiconductor device 10 in a direction orthogonal to the z direction can be reduced.

[0175] (2) For example, when the exterior electrode 70 is formed on the back surface 22 of the support layer 20, the via conductor extends in the z direction from the main surfaces 31A and 31B of the respective switching elements 30A and 30B, then extends to a position where it does not overlap with the respective switching elements 30A and 30B in the z direction, and is provided so as to penetrate the resin layer 50 and the support layer 20 in the z direction up to the back surface 22 of the support. In this way, since the via conductor extends to one side in the z direction from the main surfaces 31A and 31B of the respective switching elements 30A and 30B and then extends toward the other side in the z direction, the length of the via conductor becomes long. Therefore, the conductive path between each of the switching elements 30A and 30B and the exterior electrode 70 becomes long.

[0176] In this regard, in the present embodiment, the semiconductor device 10 includes a support layer 20 that supports the respective switching elements 30A and 30B and the driver 40. The resin layer 50 is formed on the support layer 20. The exterior electrode 70 is formed on the surface (resin main surface 51) of the resin layer 50 on the side opposite to the support layer 20 in the z direction. As a result, since the via conductor 60 is configured to extend from the main surfaces 31A and 31B of the respective switching elements 30A and 30B toward one side in the z direction toward the resin main surface 51, the length of the via conductor 60 becomes short. Therefore, the conductive path between each of the switching elements 30A and 30B and the exterior electrode 70 becomes short.

[0177] (3) The power supply via conductor 61 extends in the z direction and is disposed at a position that overlaps both the power supply electrode 71 and the first switching element 30A when viewed from the z direction. The ground via conductor 62 extends in the z direction and is disposed at a position that overlaps both the ground electrode 72 and the second switching element 30B when viewed from the z direction.

[0178] According to this configuration, since the power supply via conductor 61 extending along the z direction can connect the power supply electrode 71 and the first drive pad electrode 31AA of the first switching element 30A, the length of the conductive path between the power supply electrode 71 and the first drive pad electrode 31AA can be shortened as compared with the case where the power supply via conductor 61 is, for example, crank-shaped. Also, since the ground via conductor 62 extending along the z direction can connect the ground electrode 72 and the second drive pad electrode 31BB of the second switching element 30B, the length of the conductive path between the ground electrode 72 and the second drive pad electrode 31BB can be shortened as compared with the case where the ground via conductor 62 is, for example, crank-shaped. Therefore, the inductance caused by the lengths of these conductive paths can be reduced.

[0179] (4) The semiconductor device 10 includes an output electrode 73 as an external electrode, and an output via conductor 63 that electrically connects the output electrode 73 and each of the switching elements 30A and 30B. The output electrode 73 is disposed between the first switching element 30A and the second switching element 30B in the x direction when viewed from the z direction.

[0180] According to this configuration, both the conductive path between the first switching element 30A and the output electrode 73 and the conductive path between the second switching element 30B and the output electrode 73 can be shortened. Therefore, the inductance caused by these conductive paths can be reduced.

[0181] (5) Both the first output via conductor 63P and the second output via conductor 63Q as the output via conductor 63 are crank-shaped. According to this configuration, the first switching element 30A and the output electrode 73 can be connected, and the second switching element 30B and the output electrode 73 can be connected with a simple configuration.

[0182] (8) The first output via conductor 63P and the second output via conductor 63Q are arranged to be spaced apart from each other in the x direction. According to this configuration, the first output via conductor 63P and the second output via conductor 63Q are not directly electrically connected, but are indirectly electrically connected via the output electrode 73. Therefore, it is possible to avoid a direct current flowing from the second drive pad electrode 31AB of the first switching element 30A to the first drive pad electrode 31BA of the second switching element 30B.

[0183] (7) The resin layer 50 has an element sealing layer 50A that seals each of the switching elements 30A and 30B, and a surface-side resin layer 50B formed on the element sealing layer 50A. The first output via conductor 63P includes a first element connection via 63pa electrically connected to the first switching element 30A, a first electrode connection via 63pb connected to the output electrode 73, and a via connection wiring 63pc that connects the first element connection via 63pa and the first electrode connection via 63pb. The second output via conductor 63Q includes a second element connection via 63qa electrically connected to the second switching element 30B, a second electrode connection via 63qb connected to the output electrode 73, and a second via connection wiring 63qc that connects the second element connection via 63qa and the second electrode connection via 63qb. Each of the element connection vias 63pa and 63qa penetrates the element sealing layer 50A in the z direction. Each of the electrode connection vias 63pb and 63qb penetrates the surface-side resin layer 50B in the z direction. Each of the via connection wirings 63pc and 63qc is provided on the element sealing layer 50A. According to this configuration, each of the crank-shaped output via conductors 63P and 63Q can be easily formed.

[0184] (8) The semiconductor device 10 includes an element control via conductor 64 that electrically connects each of the switching elements 30A and 30B and the driver 40. The element control via conductor 64 is embedded in the resin layer 50.

[0185] According to this configuration, as compared with the case where each switching element 30A, 30B and the driver 40 are connected by the wire W, both the conduction path between the first switching element 30A and the driver 40 and the conduction path between the second switching element 30B and the driver 40 can be shortened. Therefore, the inductance caused by the length of these conduction paths can be reduced.

[0186] (9) The first control via conductor 64P includes a first driver-side control via 64pb electrically connected to the driver 40, a first element-side control via 64pa connected to the control pad electrode 31AC of the first switching element 30A, and a first control connection wiring 64pc connecting the first driver-side control via 64pb and the first element-side control via 64pa. The second control via conductor 64Q includes a second driver-side control via 64qb electrically connected to the driver 40, a second element-side control via 64qa connected to the control pad electrode 31BC of the second switching element 30B, and a second control connection wiring 64qc connecting the second driver-side control via 64qb and the second element-side control via 64qa. Each of the driver-side control vias 64pb, 64qb penetrates in the z direction through a portion of the element encapsulation layer 50A that covers the driver main surface 41 of the driver 40. Each of the element-side control vias 64pa, 64qa penetrates in the z direction through a portion of the element encapsulation layer 50A that covers the control pad electrodes 31AC, 31BC of the respective switching elements 30A, 30B. Each of the control connection wirings 64pc, 64qc is provided on the element encapsulation layer 50A. According to this configuration, each control via conductor 64P, 64Q can be easily formed.

[0187] (10) The semiconductor device 10 includes a driver via conductor 65 that electrically connects the driver connection electrode 74 and the driver 40. The driver via conductor 65 is provided in the resin layer 50. According to this configuration, as compared with the case where the driver connection electrode 74 and the driver 40 are connected by the wire W, the conduction path between the driver connection electrode 74 and the driver 40 can be shortened.

[0188] (11) The via conductor 65 for the driver is crank-shaped. According to this configuration, the driver pad electrode 43 of the driver 40 and the driver connection electrode 74 can be connected with a simple configuration.

[0189] (12) The first element connection via 63pa of the first output via conductor 63P is provided so as to connect the interface 57 between the element sealing layer 50A and the surface-side resin layer 50B and the second drive pad electrode 31AB of the first switching element 30A. The second element connection via 63qa of the second output via conductor 63Q is provided so as to connect the interface 57 and the first drive pad electrode 31BA of the second switching element 30B.

[0190] According to this configuration, in the manufacturing method of the semiconductor device 10, the thickness of the portion of the element sealing layer 50A that covers the element main surfaces 31A and 31B of the respective switching elements 30A and 30B is made thin. And since the respective element connection vias 63pa and 63qa are provided so as to penetrate that portion, the lengths of the respective element connection vias 63pa and 63qa can be shortened.

[0191] (13) The first element-side control via 64pa of the first control via conductor 64P is provided so as to connect the interface 57 between the element sealing layer 50A and the surface-side resin layer 50B and the control pad electrode 31AC of the first switching element 30A. The first driver-side control via 64pb is provided so as to connect the interface 57 and the driver pad electrode 43 of the driver 40. Also, the second element-side control via 64qa of the second control via conductor 64Q is provided so as to connect the interface 57 and the control pad electrode 31BC of the second switching element 30B. The second driver-side control via 64qb is provided so as to connect the interface 57 and the driver pad electrode 43 of the driver 40.

[0192] According to this configuration, in the method for manufacturing the semiconductor device 10, the thicknesses of the portions of the element sealing layer 50A that cover the element main surfaces 31A and 31B of the respective switching elements 30A and 30B and the portion that covers the driver main surface 41 of the driver 40 are made thinner. And since the element-side control vias 64pa and 64qa and the driver-side control vias 64pb and 64qb are provided so as to penetrate these portions, the lengths of the respective element-side control vias 64pa and 64qa and the respective driver-side control vias 64pb and 64qb can be shortened.

[0193] (14) The driver connection via 65a of the driver via conductor 65 is provided so as to connect the interface 57 between the element sealing layer 50A and the surface-side resin layer 50B and the driver pad electrode 43 of the driver 40. According to this configuration, in the method for manufacturing the semiconductor device 10, the thickness of the portion of the element sealing layer 50A that covers the driver main surface 41 of the driver 40 is made thinner. And since the driver connection via 65a is provided so as to penetrate that portion, the length of the driver connection via 65a can be shortened.

[0194] (15) A plurality of power supply via conductors 61 are provided. According to this configuration, the inductance of the conductive path between the first drive pad electrode 31AA of the first switching element 30A and the power supply electrode 71 can be reduced.

[0195] (16) A plurality of ground via conductors 62 are provided. According to this configuration, the inductance of the conductive path between the second drive pad electrode 31BB of the second switching element 30B and the ground electrode 72 can be reduced.

[0196] (17) A plurality of element connection vias 63pa and 63qa of the respective output via conductors 63P and 63Q are provided. According to this configuration, both the inductance of the conductive path between the second drive pad electrode 31AB of the first switching element 30A and the first via connection wiring 63pc and the inductance of the conductive path between the first drive pad electrode 31BA of the second switching element 30B and the second via connection wiring 63qc can be reduced.

[0197] (18) A plurality of via electrodes 63pb and 63qb for electrode connection of each output via conductor 63P and 63Q are provided. According to this configuration, both the inductance of the conductive path between the first via connection wiring 63pc and the output electrode 73 and the inductance of the conductive path between the second via connection wiring 63qc and the output electrode 73 can be reduced.

[0198] (19) Each of the via connection wirings 63pc and 63qc, each of the control connection wirings 64pc and 64qc, and the via connection wiring 65c is disposed on the element sealing layer 50A. According to this configuration, these wirings 63pc, 63qc, 64pc, 64qc, and 65c can be formed in the same process. Therefore, the manufacturing process of the semiconductor device 10 can be simplified.

[0199] (20) The manufacturing method of the semiconductor device 10 includes a step of forming an element sealing layer 850A that seals each of the switching elements 30A and 30B, a step of forming a surface-side resin layer 850B on the element sealing layer 850A, a step of forming through holes 58a and 58b in the element sealing layer 850A and the surface-side resin layer 850B so that each of the switching elements 30A and 30B is exposed, a step of embedding a power supply via conductor 61 in the through hole 58a and embedding a ground via conductor 62 in the through hole 58b, and a step of forming a power supply electrode 71 and a ground electrode 72 on the surface-side resin layer 850B. According to this configuration, an effect similar to the effect of the above (1) can be obtained.

[0200] (21) The manufacturing method of the semiconductor device 10 includes a grinding process for removing the element sealing layer 850A in the z direction. According to this configuration, the thickness of the portion of the element sealing layer 850A that covers each element main surface 31A, 31B of each switching element 30A, 30B in the z direction can be made thinner than the thickness of the portion of the element sealing layer 850A that covers each element main surface 31A, 31B in the z direction when the grinding process is not provided. Also, the thickness of the portion of the element sealing layer 850A that covers the driver main surface 41 of the driver 40 in the z direction can be made thinner than the thickness of the portion of the element sealing layer 850A that covers the driver main surface 41 in the z direction when the grinding process is not provided. And each element connection via 63pa, 63qa and each element side control via 64pa, 64qa penetrate the portion of the element sealing layer 850A that covers each element main surface 31A, 31B in the z direction in the z direction, and each driver side control via 64pb, 64qb and driver connection via 65a penetrate the portion of the element sealing layer 850A that covers the driver main surface 41 in the z direction in the z direction. Thereby, the distance in the z direction between each switching element 30A, 30B and each via connection wiring 63pc, 63qc and each control connection wiring 64pc, 64qc, and the distance in the z direction between the driver 40 and each control connection wiring 64pc, 64qc and via connection wiring 65c can be shortened. Therefore, the conductive path between the second drive pad electrode 31AB of the first switching element 30A and the first via connection wiring 63pc, the conductive path between the first drive pad electrode 31BA of the second switching element 30B and the second via connection wiring 63qc, the conductive path between the control pad electrode 31AC of the first switching element 30A and the first control connection wiring 64pc, the conductive path between the control pad electrode 31BC of the second switching element 30B and the second control connection wiring 64qc, and the conductive path between the driver pad electrode 43 of the driver 40 and the via connection wiring 65c can each be shortened.

[0201] (Modified Example) The above-described embodiments are examples of forms that the semiconductor device according to the present disclosure can take, and are not intended to limit the form. The semiconductor device according to the present disclosure can take a form different from the form exemplified in the above-described embodiments. An example thereof is a form in which a part of the configuration of the above-described embodiment is replaced, changed, or omitted, or a form in which a new configuration is added to the above-described embodiment. Also, the following modification examples can be combined with each other as long as there is no technical contradiction. In the following modification examples, parts common to the above-described embodiments are denoted by the same reference numerals as in the above-described embodiments, and the description thereof is omitted.

[0202] · In the above-described embodiment, the semiconductor device 10 had two switching elements, i.e., the first switching element 30A and the second switching element 30B, as the switching element 30, but it is not limited thereto. For example, as shown in FIGS. 27 to 30, the semiconductor device 10 may include one switching element 30. In FIG. 27, the resin layer 50 is omitted for convenience of explanation. Also, in FIG. 28, the surface-side resin layer 50B is omitted for convenience of explanation. In the semiconductor device 10 of the modification example shown in FIGS. 27 to 30, the driver 40 corresponds to the specific element.

[0203] As shown in FIG. 27, the semiconductor device 10 of the modification example shown in FIGS. 27 to 30 has a smaller size in the x direction compared to the semiconductor device 10 of the above-described embodiment because the number of switching elements has decreased.

[0204] The switching element 30 is disposed at the center in the x direction and closer to the support side surface 24 than the support side surface 23 in the y direction on the support main surface 21 of the support layer 20. The switching element 30 has, for example, a GaN HEMT as in the above-described embodiment. The shape of the switching element 30 viewed from the z direction is a rectangular shape having a long side direction and a short side direction. In the illustrated example, the switching element 30 is disposed on the support main surface 21 such that the long side direction is along the y direction and the short side direction is along the x direction.

[0205] On the main element surface 31 of the switching element 30, a first drive pad electrode 33A, a second drive pad electrode 33B, and a control pad electrode 33C are formed. In the illustrated example, the first drive pad electrode 33A constitutes a drain electrode, the second drive pad electrode 33B constitutes a source electrode, and the control pad electrode 33C constitutes a gate electrode. The arrangement configuration of these electrodes 33A to 33C is the same as the arrangement configuration of the first drive pad electrode 31AA, the second drive pad electrode 31AB, and the control pad electrode 31AC of the first switching element 30A in the above embodiment.

[0206] The driver 40 is mounted on the main support surface 21. That is, the driver 40 is arranged on the same plane as the switching element 30. The driver 40 is arranged near the support side surface 23 rather than the switching element 30 in the y direction and at the center in the x direction on the main support surface 21.

[0207] As shown in FIG. 29, on the resin main surface 51 of the element sealing layer 50A, a power supply electrode 71, a ground electrode 72, and a plurality of driver connection electrodes 74 are formed as external electrodes 70. The configuration of the plating layer of the external electrode 70 in the illustrated example is the same as the configuration of the plating layer of the external electrode 70 in the above embodiment. Also, the external electrode 70 in the illustrated example does not have the output electrode 73 in the above embodiment.

[0208] As shown in FIG. 29, the power supply electrode 71 and the ground electrode 72 are arranged spaced apart from each other in the x direction while being aligned with each other in the y direction. When viewed from the z direction, the power supply electrode 71 and the ground electrode 72 are arranged at positions overlapping the switching element 30. When viewed from the z direction, the power supply electrode 71 has an overhanging portion protruding in the x direction with respect to the switching element 30. When viewed from the z direction, the overhanging portion of the power supply electrode 71 extends from the switching element 30 toward the device side surface 15. The ground electrode 72 has an overhanging portion protruding in the x direction with respect to the switching element 30. When viewed from the z direction, the overhanging portion of the ground electrode 72 extends from the switching element 30 toward the device side surface 16.

[0209] The power supply via conductor 61 is disposed, as in the above-described embodiment, at a position overlapping both the power supply electrode 71 and the first drive pad electrode 33A of the switching element 30 when viewed from the z direction, and connects the power supply electrode 71 and the first drive pad electrode 33A.

[0210] As shown in FIG. 30, the power supply via conductor 61 extends along the z direction. More specifically, the power supply via conductor 61 penetrates, as in the above-described embodiment, the surface-side resin layer 50B and the portion of the element sealing layer 50A that covers the element main surface 31 of the switching element 30 in the z direction.

[0211] As shown in FIG. 29, the ground via conductor 62 is disposed, as in the above-described embodiment, at a position overlapping both the ground electrode 72 and the second drive pad electrode 33B of the switching element 30 when viewed from the z direction, and connects the ground electrode 72 and the second drive pad electrode 33B.

[0212] As shown in FIG. 30, the ground via conductor 62 extends along the z direction. More specifically, the ground via conductor 62 penetrates, as in the above-described embodiment, the surface-side resin layer 50B and the portion of the element sealing layer 50A that covers the element main surface 31 of the switching element 30 in the z direction.

[0213] As shown in FIG. 28, the semiconductor device 10 includes an element control via conductor 64. The element control via conductor 64 connects the control pad electrode 33C of the switching element 30 and the driver pad electrode 43 of the driver 40. The element control via conductor 64 includes an element-side control via 64a, a driver-side control via 64b, and a control connection wiring 64c.

[0214] The element-side control via 64a is disposed, when viewed from the z direction, at a position overlapping the control pad electrode 33C, and penetrates the portion of the element sealing layer 50A that covers the element main surface 31 of the switching element 30 in the z direction.

[0215] The driver-side control via 64b is arranged at a position overlapping with the driver pad electrode 43 of the driver 40 when viewed from the z direction, and penetrates in the z direction through a portion of the element sealing layer 50A that covers the driver main surface 41 of the driver 40. The driver pad electrode 43 connected to the driver-side control via 64b is arranged at an end portion closer to the switching element 30 among both end portions in the y direction of the driver main surface 41 and at a position facing the control pad electrode 33C in the y direction.

[0216] The control connection wiring 64c is a wiring connecting the element-side control via 64a and the driver-side control via 64b, and is formed on the element sealing layer 50A. When viewed from the z direction, the control connection wiring 64c is strip-shaped and extends along the y direction. Among both end portions in the y direction of the control connection wiring 64c, the end portion closer to the switching element 30 is provided so as to overlap with the element-side control via 64a when viewed from the z direction. Among both end portions in the y direction of the control connection wiring 64c, the end portion closer to the driver 40 is provided so as to overlap with the driver-side control via 64b when viewed from the z direction. Therefore, it can also be said that the element-side control via 64a is arranged at a position overlapping with both the control pad electrode 33C and the control connection wiring 64c when viewed from the z direction. The driver-side control via 64b is arranged at a position overlapping with both the driver pad electrode 43 and the control connection wiring 64c when viewed from the z direction.

[0217] As shown in FIG. 28, the configuration of the driver via conductor 65 connecting the driver pad electrode 43 and the driver connection electrode 74 of the driver 40 is the same as the configuration of the driver via conductor 65 of the above embodiment. However, in the illustrated example, since the positional relationship between the driver pad electrode 43 and the driver connection electrode 74 is different from the positional relationship between the driver pad electrode 43 and the driver connection electrode 74 of the above embodiment, the shape of the via connection wiring 65c when viewed from the z direction is different from the shape of the via connection wiring 65c of the above embodiment.

[0218] The manufacturing method of the semiconductor device 10 of the modified example shown in FIGS. 27 to 30 is generally the same as the manufacturing method of the above embodiment. Specifically, the manufacturing method of the semiconductor device 10 of the modified example mainly includes a support layer formation step, an element mounting step, an element sealing layer formation step, a first via conductor formation step, a surface-side resin layer formation step, a second via conductor formation step, an external electrode formation step, and a dicing step. The support layer formation step is the same as the support layer formation step of the above embodiment.

[0219] The element mounting step is different in that the switching element 30 is mounted on the base material 820 instead of the switching elements 30A and 30B. The mounting method of the switching element 30 is the same as the element mounting step of the above embodiment. The element sealing layer formation step is the same as the element sealing layer formation step of the above embodiment.

[0220] In the first via conductor formation step, the element-side control via 64a, the driver-side control via 64b, the control connection wiring 64c, the plurality of driver connection vias 65a, and the plurality of via connection wirings 65c are formed. In other words, in the first via conductor formation step, the element connection vias 63pa and 63qa, the via connection wirings 63pc and 63qc, the element-side control vias 64pa and 64qa, the driver-side control vias 64pb and 64qb, and the control connection wirings 64pc and 64qc are not formed. The formation method of the element-side control via 64a is, for example, the same as the formation method of the first element-side control via 64pa of the above embodiment. The formation method of the driver-side control via 64b is, for example, the same as the formation method of the first driver-side control via 64pb of the above embodiment. The formation method of the control connection wiring 64c is, for example, the same as the formation method of the first control connection wiring 64pc of the above embodiment. The formation method of the plurality of driver connection vias 65a is the same as the formation method of the driver connection vias 65a of the above embodiment. The formation method of the plurality of via connection wirings 65c is the same as the formation method of the via connection wirings 65c of the above embodiment. Also, the surface-side resin layer formation step is the same as the surface-side resin layer formation step of the above embodiment.

[0221] In the second via conductor formation step, a power supply via conductor 61, a ground via conductor 62, and a plurality of driver electrode connection vias 65b are formed. In other words, in the second via conductor formation step, each electrode connection via 63pb, 63qb is not formed. The method for forming the power supply via conductor 61, the ground via conductor 62, and the plurality of driver electrode connection vias 65b is the same as the method for forming the power supply via conductor 61, the ground via conductor 62, and the plurality of driver electrode connection vias 65b in the above embodiment.

[0222] The external electrode formation step is different from the external electrode formation step of the above embodiment in that the output electrode 73 is not formed. The method for forming the power supply electrode 71, the ground electrode 72, and the plurality of driver connection electrodes 74 is the same as that of the above embodiment. The cutting step is the same as the cutting step of the above embodiment.

[0223] · In the above embodiment, the driver 40 may be omitted. In one example, as shown in FIGS. 31 to 34, the semiconductor device 10 includes a first switching element 30A and a second switching element 30B. The arrangement relationship between the first switching element 30A and the second switching element 30B is the same as that of the above embodiment. In FIG. 31, for convenience of explanation, the resin layer 50 is omitted. In FIG. 32, for convenience of explanation, the surface-side resin layer 50B is omitted. In the semiconductor device 10 of the modified example shown in FIGS. 31 to 34, the second switching element 30B corresponds to the specific element.

[0224] As shown in FIG. 31, with the omission of the driver 40, each control via conductor 64P, 64Q and the plurality of driver vias 65 are omitted. Also, due to the omission of the driver 40, each control via conductor 64P, 64Q and the plurality of driver vias 65, the semiconductor device 10 of the modified example shown in FIGS. 31 to 34 has a smaller size in the y direction compared to the semiconductor device 10 of the above embodiment.

[0225] As shown in FIGS. 32 and 33, the power via conductor 61, the ground via conductor 62, and the output via conductor 63 are the same as those in the above-described embodiment. On the other hand, the configurations of the first control via conductor 64P and the second control via conductor 64Q are different from those in the above-described embodiment.

[0226] As shown in FIGS. 33 and 34, the exterior electrode 70 is formed on the resin main surface 51 of the resin layer 50 and has a power electrode 71, a ground electrode 72, an output electrode 73, a first control electrode 75, and a second control electrode 76. The arrangement positions of the power electrode 71, the ground electrode 72, and the output electrode 73 on the resin main surface 51 are the same as those in the above-described embodiment.

[0227] The first control electrode 75 is disposed at a position overlapping with the control pad electrode 31AC of the first switching element 30A when viewed from the z direction. The second control electrode 76 is disposed at a position overlapping with the control pad electrode 31BC of the second switching element 30B when viewed from the z direction. Each of the control electrodes 75 and 76 is disposed closer to the resin side surface 53 than the power electrode 71, the ground electrode 72, and the output electrode 73 on the resin main surface 51 in the y direction. The control electrodes 75 and 76 are arranged spaced apart from each other in the x direction while being aligned with each other in the y direction. Each of the control electrodes 75 and 76 includes a Cu layer and a plating layer covering the Cu layer, similar to the power electrode 71 and the like. The Cu layer includes a seed layer formed on the resin main surface 51 and a plating layer formed on the seed layer. The seed layer is made of, for example, Cu or Ti. The plating layer is made of Cu. The plating layer covering the Cu layer is composed of a laminate of Ni, Pd, and Au. Each of the control electrodes 75 and 76 is a terminal for electrically connecting to a driver provided outside the semiconductor device 10. Here, the driver is a circuit that controls each of the switching elements 30A and 30B, similar to the driver 40 in the above-described embodiment.

[0228] As shown in FIGS. 33 and 34, the first control via conductor 64P is a conductor that connects the control pad electrode 31AC of the first switching element 30A and the first control electrode 75. The first control via conductor 64P is disposed at a position overlapping both the control pad electrode 31AC and the first control electrode 75 when viewed from the z direction.

[0229] As shown in FIG. 34, the first control via conductor 64P is provided so as to penetrate in the z direction through the surface-side resin layer 50B and the portion of the element sealing layer 50A that covers the first element main surface 31A of the first switching element 30A. The first control via conductor 64P extends along the z direction. The length of the first control via conductor 64P in the z direction is equal to the length of the power supply via conductor 61 in the z direction. Here, if the difference between the length of the first control via conductor 64P in the z direction and the length of the power supply via conductor 61 in the z direction is, for example, 10% or less of the length of the first control via conductor 64P in the z direction, it can be said that the length of the first control via conductor 64P in the z direction is equal to the length of the power supply via conductor 61 in the z direction.

[0230] As shown in FIGS. 33 and 34, the second control via conductor 64Q is a conductor that connects the control pad electrode 31BC of the second switching element 30B and the second control electrode 76. The second control via conductor 64Q is disposed at a position overlapping both the control pad electrode 31BC and the second control electrode 76 when viewed from the z direction.

[0231] As shown in FIG. 34, the second control via conductor 64Q is provided so as to penetrate in the z direction through the surface-side resin layer 50B and the portion of the element sealing layer 50A that covers the second element main surface 31B of the second switching element 30B. The second control via conductor 64Q extends along the z direction. The length of the second control via conductor 64Q in the z direction is equal to the length of the ground via conductor 62 in the z direction. Here, if the difference between the length of the second control via conductor 64Q in the z direction and the length of the ground via conductor 62 in the z direction is, for example, 10% or less of the length of the second control via conductor 64Q in the z direction, it can be said that the length of the second control via conductor 64Q in the z direction is equal to the length of the ground via conductor 62 in the z direction.

[0232] The manufacturing method of the semiconductor device 10 according to the modified example shown in FIGS. 31 to 34 is generally the same as the manufacturing method of the above embodiment. Specifically, the manufacturing method of the semiconductor device 10 according to the modified example mainly includes a support layer forming step, an element mounting step, an element sealing layer forming step, a first via conductor forming step, a surface side resin layer forming step, a second via conductor forming step, an external electrode forming step, and a dicing step. The support layer forming step is the same as the support layer forming step of the above embodiment.

[0233] The element mounting step is different in that each of the switching elements 30A and 30B is mounted on the base material 820 and the driver 40 is not mounted. The method for each of the switching elements 30A and 30B is the same as the element mounting step of the above embodiment. The element sealing layer forming step is the same as the element sealing layer forming step of the above embodiment.

[0234] In the first via conductor forming step, each element side control via 64pa, 64qa and each control connection wiring 64pc, 64qc are formed. In other words, in the first via conductor forming step, each control via conductor 64P, 64Q is not formed. The forming method of each element side control via 64pa, 64qa is the same as the forming method of each element side control via 64pa, 64qa of the above embodiment. The forming method of each control connection wiring 64pc, 64qc is the same as the forming method of each control connection wiring 64pc, 64qc of the above embodiment. Also, the surface side resin layer forming step is the same as the surface side resin layer forming step of the above embodiment.

[0235] In the second via conductor forming step, a power supply via conductor 61, a ground via conductor 62, each electrode connection via 63pb, 63qb, a first control via conductor 64P, and a second control via conductor 64Q are formed. In other words, in the second via conductor forming step, the driver electrode connection via 65b is not formed. The forming methods of the power supply via conductor 61, the ground via conductor 62, and each electrode connection via 63pb, 63qb are the same as the forming methods of the power supply via conductor 61, the ground via conductor 62, and each electrode connection via 63pb, 63qb of the above embodiment. The forming methods of the first control via conductor 64P and the second control via conductor 64Q include a through hole forming step and a via forming step.

[0236] In the through-hole forming step, through-holes 58m and 58n are formed by drilling such as laser processing. The through-hole 58m is provided at a position overlapping with the control pad electrode 31AC of the first switching element 30A when viewed from the z direction. The through-hole 58m penetrates in the z direction through a portion of the surface-side resin layer 50B and the element sealing layer 50A that covers the first element main surface 31A of the first switching element 30A. Therefore, through the through-hole 58m, the control pad electrode 31AC is exposed from the resin layer 50 in the z direction. The through-hole 58n is provided at a position overlapping with the control pad electrode 31BC of the second switching element 30B when viewed from the z direction. The through-hole 58n penetrates in the z direction through a portion of the surface-side resin layer 50B and the element sealing layer 50A that covers the second element main surface 31B of the second switching element 30B. Therefore, through the through-hole 58n, the control pad electrode 31BC is exposed from the resin layer 50 in the z direction.

[0237] In the via forming step, the first control via conductor 64P and the second control via conductor 64Q are formed by embedding metal vias in the through-holes 58m and 58n, respectively. The via forming step for each control via conductor 64P, 64Q is the same as the via forming step when forming, for example, the power supply via conductor 61.

[0238] The exterior electrode forming step is different from the exterior electrode forming step of the above embodiment in that each control electrode 75, 76 is formed and a plurality of driver connection electrodes 74 are not formed. The forming methods of the power supply electrode 71, the ground electrode 72, and the output electrode 73 are the same as those of the above embodiment. The first control electrode 75 is formed so as to cover the first control via conductor 64P from the z direction. The second control electrode 76 is formed so as to cover the second control via conductor 64Q from the z direction. The forming methods of each control electrode 75, 76 are the same as the forming method of the power supply electrode 71 or the like, and are formed by electroless plating. The cutting step is the same as the cutting step of the above embodiment.

[0239] · In the above embodiment, metal layers 81 to 83 made of metal may be provided on the support layer 20. In one example, as shown in FIG. 35, the metal layer 81 supports the first switching element 30A. In other words, the first switching element 30A is mounted on the metal layer 81. The first switching element 30A is joined to the metal layer 81 by a joining material SD. The metal layer 82 supports the second switching element 30B. In other words, the second switching element 30B is mounted on the metal layer 82. The second switching element 30B is joined to the metal layer 82 by a joining material SD. The metal layer 83 supports the driver 40. In other words, the driver 40 is mounted on the metal layer 83. The driver 40 is joined to the metal layer 83 by a joining material SD. The arrangement relationship and size of each of the switching elements 30A, 30B and the driver 40 are the same as those in the above embodiment. For this reason, the metal layer 81 and the metal layer 82 are arranged to be spaced apart from each other in the x direction while being aligned with each other in the y direction. The metal layer 83 is arranged closer to the support side surface 23 than the metal layers 81 and 82 in the y direction in the support layer 20. The metal layer 83 is arranged at the center of the support layer 20 in the x direction. In the illustrated example, when viewed from the z direction, the metal layer 81 is formed to have a size slightly larger than the first element main surface 31A of the first switching element 30A. The metal layer 82 is formed to have a size slightly larger than the second element main surface 31B of the second switching element 30B. The metal layer 83 is formed to have a size slightly larger than the driver main surface 41 of the driver 40. The metal layers 81 to 83 include, for example, a laminate of Ni, Pd, and Au.

[0240] As shown in FIG. 36, the metal layers 81 and 82 are provided so as to penetrate the support layer 20 in the z direction. Although not shown, the metal layer 83 is provided so as to penetrate the support layer 20 in the z direction in the same manner as the metal layers 81 and 82. The portion of the support layer 20 other than the metal layers 81 to 82 is made of the same material as the support layer 20 in the above embodiment, that is, an electrically insulating material. An epoxy resin is used as the electrically insulating material. That is, the support layer 20 has an insulating layer 85 that electrically insulates the metal layers 81 to 83.

[0241] As described above, the support layer 20 includes a metal layer 81 which is a first metal layer on which a first switching element 30A, which is an example of a switching element, is mounted, a metal layer 82 which is a second metal layer on which a second switching element 30B, which is an example of a specific element, is mounted, and an insulating layer 85 which electrically insulates the metal layer 81 and the metal layer 82. Note that the specific element is not limited to the second switching element 30B and may be a driver 40. In this case, the second metal layer becomes a metal layer 83.

[0242] According to the semiconductor device 10 of the modified example shown in FIGS. 35 and 36, the heat of each of the switching elements 30A and 30B is released from the metal layers 81 and 82 to the outside of the semiconductor device 10. Therefore, it is possible to improve the heat dissipation performance of each of the switching elements 30A and 30B. Further, the heat of the driver 40 is released from the metal layer 83 to the outside of the semiconductor device 10. Therefore, it is possible to improve the heat dissipation performance of the driver 40.

[0243] The manufacturing method of the semiconductor device 10 of the modified example shown in FIGS. 35 and 36 is substantially the same as the manufacturing method of the above-described embodiment. Specifically, the manufacturing method of the semiconductor device 10 of the modified example mainly includes a support layer forming step, an element mounting step, an element sealing layer forming step, a first via conductor forming step, a surface-side resin layer forming step, a second via conductor forming step, an external electrode forming step, and a cutting step. The manufacturing method of the semiconductor device 10 of the modified example is different from the above-described embodiment in the external electrode forming step.

[0244] Specifically, the external electrode forming step in the manufacturing method of the semiconductor device 10 of the modified example includes a base material processing step and a plating layer forming step. In the base material processing step, through holes for forming the metal layers 81 to 83 are formed in portions of the base material 820 corresponding to each of the switching elements 30A and 30B and the driver 40. The through holes are formed, for example, by laser processing. In the plating layer forming step, the external electrodes 70 and the metal layers 81 to 83 are formed. Each of the external electrodes 70 and the metal layers 81 to 83 is formed, for example, by electroless plating.

[0245] Note that the method for forming the metal layers 81 to 83 is not limited to the electroless plating method and can be arbitrarily changed. In one example, the metal layers 81 to 83 may be formed using a pre-formed metal plate. An example of the metal plate is a copper plate or an aluminum plate. In this case, the metal layers 81 to 83 are formed in the support layer forming step instead of the exterior electrode forming step. Specifically, in the support layer forming step, first, the metal layers 81 to 83 are disposed on the substrate main surface 801 of the support substrate 800. Subsequently, the base material 820 is formed on the substrate main surface 801.

[0246] · In the semiconductor device 10 of the modification example shown in FIGS. 35 and 36, the second drive pad electrode 31BB of the second switching element 30B may be electrically connected to the metal layer 82. In one example, as shown in FIG. 37, the length in the x direction of the protruding portion of the metal layer 82 that protrudes toward the support side surface 26 with respect to the second switching element 30B is longer than the length in the x direction of the protruding portion of the metal layer 82 that protrudes toward the support side surface 25 with respect to the second switching element 30B. The semiconductor device 10 includes a connection via conductor 84 that connects the ground electrode 72 and the metal layer 82. The connection via conductor 84 is provided at a position that overlaps both the ground electrode 72 and the metal layer 82 when viewed from the z direction and does not overlap the second switching element 30B. More specifically, the connection via conductor 84 is provided adjacent to the device side surface 16 side with respect to the second switching element 30B in the x direction. The connection via conductor 84 is provided so as to penetrate the resin layer 50. Thereby, the connection via conductor 84 is in contact with both the ground electrode 72 and the metal layer 82.

[0247] According to this configuration, since the metal layer 82 is electrically connected to the ground electrode 72 by the connection via conductor 84, the metal layer 82 also becomes the ground. Therefore, the influence of noise on each of the switching elements 30A and 30B can be reduced.

[0248] · In the above embodiment, the driver 40 may not be arranged on the same plane with respect to each of the switching elements 30A and 30B. In one example, as shown in FIGS. 38 and 39, the driver 40 is arranged at a position shifted in the z direction with respect to each of the switching elements 30A and 30B. More specifically, as shown in FIG. 39, the driver 40 is arranged on the element sealing layer 50A. As shown in FIG. 38, when viewed from the z direction, the driver 40 is arranged at a position overlapping the switching elements 30A and 30B. More specifically, when viewed from the z direction, the driver 40 is arranged on the element sealing layer 50A such that a predetermined driver pad electrode 43 of the driver 40 overlaps with the control pad electrode 31AC of the first switching element 30A, and another driver pad electrode 43 overlaps with the control pad electrode 31BC of the second switching element 30B. In the illustrated example, the driver 40 is arranged such that the driver main surface 41 faces the device back surface 12 side and the driver back surface 42 faces the device main surface 11 side.

[0249] The configurations of the first control via conductor 64P and the second control via conductor 64Q are different from the configurations of the respective control via conductors 64P and 64Q of the above embodiment.

[0250] The first control via conductor 64P is arranged at a position overlapping both the control pad electrode 31AC of the first switching element 30A and a predetermined driver pad electrode 43 of the driver 40 when viewed from the z direction. The first control via conductor 64P has a first element side control via 64pa and a first control connection wiring 64pc. In other words, the first control via conductor 64P does not have a first driver side control via 64pb. The first element side control via 64pa is provided so as to penetrate in the z direction a portion of the element encapsulation layer 50A covering the first element main surface 31A of the first switching element 30A, similar to the above-described embodiment. The first control connection wiring 64pc is provided so as to cover a portion of the first element side control via 64pa exposed from the element encapsulation layer 50A. The first control connection wiring 64pc is formed in a rectangular shape having substantially the same size as the predetermined driver pad electrode 43. Therefore, it can be said that the first control via conductor 64P extends along the z direction. The predetermined driver pad electrode 43 is joined to the first control connection wiring 64pc by a conductive bonding material made of solder or Ag paste.

[0251] The second control via conductor 64Q is arranged at a position overlapping both the control pad electrode 31BC of the second switching element 30B and another driver pad electrode 43 of the driver 40 when viewed from the z direction. Although not shown, the second control via conductor 64Q has a second element side control via 64qa and a second control connection wiring 64qc. The second element side control via 64qa is provided so as to penetrate in the z direction a portion of the element encapsulation layer 50A covering the second element main surface 31B of the second switching element 30B, similar to the above-described embodiment. The second control connection wiring 64qc is provided so as to cover a portion of the second element side control via 64qa exposed from the element encapsulation layer 50A. The second control connection wiring 64qc is formed in a rectangular shape having substantially the same size as the other driver pad electrode 43. Therefore, it can also be said that the second control via conductor 64Q extends along the z direction. The other driver pad electrode 43 is joined to the second control connection wiring 64qc by a conductive bonding material made of solder or Ag paste.

[0252] Although not shown, the configuration of the driver via conductor 65 is different from that of the driver via conductor 65 in the above embodiment. Specifically, in the driver via conductor 65 of the modification example, the driver connection via 65a is omitted. That is, the driver 40 is arranged closer to the device main surface 11 than the via connection wiring 65c in the z direction, and the driver pad electrode 43 is joined to the via connection wiring 65c by a conductive joining material made of solder or Ag paste.

[0253] According to the semiconductor device 10 of the modification example shown in FIGS. 38 and 39, since the lengths of the control via conductors 64P and 64Q are shortened, the conductive path between the control pad electrodes 31AC and 31BC of the switching elements 30A and 30B and the driver 40 is likely to be shortened. Therefore, the inductance caused by the length of this conductive path can be reduced.

[0254] · In the semiconductor device 10 of the modification example shown in FIGS. 38 and 39, each of the switching elements 30A and 30B is mounted on the support main surface 21 of the support layer 20, and the driver 40 is mounted on the element sealing layer 50A, but it is not limited to this. For example, each of the switching elements 30A and 30B may be mounted on the element sealing layer 50A, and the driver 40 may be mounted on the support main surface 21. In this case, the driver 40 is arranged such that the driver main surface 41 faces the device main surface 11 side and the driver back surface 42 faces the device back surface 12 side. Also, each of the switching elements 30A and 30B is arranged such that each element main surface 31A and 31B faces the device back surface 12 side and each element back surface 32A and 32B faces the device main surface 11 side.

[0255] · In the above embodiment, the surface-side resin layer 50B may be omitted from the resin layer 50. In one example, as shown in FIG. 40, each electrode connection via 63pb and 63qb is omitted from each output via conductor 63P and 63Q. That is, the first output via conductor 63P has a first element connection via 63pa and a first via connection wiring 63pc, and the second output via conductor 63Q has a second element connection via 63qa and a second via connection wiring 63qc. Each via connection wiring 63pc and 63qc is provided outside the resin layer 50. The output electrode 73 is formed on each via connection wiring 63pc and 63qc. That is, the first via connection wiring 63pc is a wiring that connects the first element connection via 63pa and the output electrode 73, and the second via connection wiring 63qc is a wiring that connects the second element connection via 63qa and the output electrode 73. The output electrode 73 formed in the portion between the first via connection wiring 63pc and the second via connection wiring 63qc in the x direction is formed on the element sealing layer 50A.

[0256] Although not shown, the power supply via conductor 61 penetrates in the z direction through the portion covering the first main element surface 31A of the first switching element 30A. The ground via conductor 62 penetrates in the z direction through the portion covering the second main element surface 31B of the second switching element 30B. For this reason, the length in the z direction of the power supply via conductor 61 and the ground via conductor 62 is shorter than the length in the z direction of the power supply via conductor 61 and the ground via conductor 62 in the above embodiment. Each of the power supply electrode 71 and the ground electrode 72 is formed on the element sealing layer 50A. The power supply electrode 71 is formed so as to cover the portion of the power supply via conductor 61 exposed from the element sealing layer 50A. The ground electrode 72 is formed so as to cover the portion of the ground via conductor 62 exposed from the element sealing layer 50A.

[0257] According to this configuration, the conductive paths between the second drive pad electrode 31AB of the first switching element 30A and the output electrode 73, between the first drive pad electrode 31BA of the second switching element 30B and the output electrode 73, between the first drive pad electrode 31AA of the first switching element 30A and the power supply electrode 71, and between the second drive pad electrode 31BB of the second switching element 30B and the ground electrode 72 can each be shortened. Also, the conductive path between the driver pad electrode 43 of the driver 40 and the driver connection electrode 74 can be shortened. Therefore, the inductance caused by the lengths of these conductive paths can be reduced.

[0258] · In the above embodiment, the z-direction lengths of both the power supply via conductor 61 and the ground via conductor 62 were greater than the z-direction lengths (the thicknesses of the respective switching elements 30A and 30B) of the respective switching elements 30A and 30B, but it is not limited to this. The z-direction lengths of both the power supply via conductor 61 and the ground via conductor 62 can be arbitrarily changed. In one example, the z-direction lengths of both the power supply via conductor 61 and the ground via conductor 62 may be equal to the z-direction lengths (the thicknesses of the respective switching elements 30A and 30B) of the respective switching elements 30A and 30B. Also, the z-direction lengths of both the power supply via conductor 61 and the ground via conductor 62 may be smaller than the z-direction lengths (the thicknesses of the respective switching elements 30A and 30B) of the respective switching elements 30A and 30B. According to this configuration, the conductive path between the first drive pad electrode 31AA of the first switching element 30A and the power supply electrode 71 and the conductive path between the second drive pad electrode 31BB of the second switching element 30B and the ground electrode 72 can each be shortened. Therefore, the inductance caused by the lengths of these conductive paths can be reduced.

[0259] · In the above embodiment, the support layer 20 may be omitted. In this case, each element back surface 32 of each of the switching elements 30A and 30B and the driver back surface 42 of the driver 40 are flush with the resin back surface 52 of the resin layer 50. In other words, each of the element back surfaces 32 of the switching elements 30A and 30B and the driver back surface 42 of the driver 40 are exposed in the z direction from the resin layer 50.

[0260] · In the above embodiment, the first switching element 30A and the second switching element 30B had the same configuration, but the present invention is not limited to this. For example, the first switching element 30A and the second switching element 30B may have different configurations from each other.

[0261] · In the above embodiment, the arrangement pattern of the power supply electrode 71, the ground electrode 72, and the output electrode 73 can be arbitrarily changed. In one example, the power supply electrode 71 may be arranged between the first switching element 30A and the second switching element 30B in the x direction, and the output electrode 73 may be arranged at a position overlapping the first switching element 30A when viewed from the z direction. That is, the arrangement positions of the power supply electrode 71 and the output electrode 73 may be interchanged.

[0262] · In the above embodiment, in the element sealing layer forming step of the manufacturing method of the semiconductor device 10, the grinding step of grinding the element sealing layer 850A in the z direction may be omitted.

[0263] · In the above embodiment, the thickness of the element sealing layer 50A can be arbitrarily changed. In one example, the thickness of the portion of the element sealing layer 50A that covers the main surfaces 31A and 31B of the switching elements 30A and 30B may be equal to the thickness of the surface-side resin layer 50B, or may be thicker than the thickness of the surface-side resin layer 50B. The thickness of the portion of the element sealing layer 50A that covers the driver main surface 41 of the driver 40 may be equal to the thickness of the surface-side resin layer 50B, or may be thicker than the thickness of the surface-side resin layer 50B.

[0264] (Supplementary Note) The technical idea that can be grasped from the above-described embodiments and the above-described modification examples will be described below.

[0265] (Appendix A1) A switching element, A specific element, A semiconductor device including a resin layer that seals the switching element and the specific element, with the thickness direction of the resin layer being the height direction, An external electrode formed on the surface of the resin layer, at least a part of which overlaps the switching element when viewed from the height direction, A semiconductor device including an element driving via conductor that penetrates the resin layer in the height direction to electrically connect the switching element and the external electrode.

[0266] (Appendix A2) The resin layer An element sealing layer that seals the first switching element and the second switching element, And a surface-side resin layer formed on the element sealing layer, The thickness of the surface-side resin layer is thinner than the thickness of the element sealing layer The semiconductor device according to Appendix A1.

[0267] (Appendix A3) The length of the element driving via conductor in the height direction is smaller than the length of the switching element in the height direction The semiconductor device according to Appendix A2.

[0268] (Appendix A4) The switching element is a first switching element, The specific element is a second switching element connected in series to the first switching element, and is arranged at an interval from the first switching element in a direction orthogonal to the height direction, An output electrode is formed on the surface of the resin layer, The semiconductor device includes an output via conductor that electrically connects the output electrode, the first switching element, and the second switching element. The external electrode has a power supply electrode and a ground electrode provided on both sides of the output electrode in the arrangement direction of the first switching element and the second switching element. The element driving via includes a power supply via conductor that electrically connects the first switching element and the power supply electrode, and a ground via conductor that electrically connects the second switching element and the ground electrode. In the arrangement direction, the output via conductor is disposed between the power supply via conductor and the ground via conductor. The semiconductor device according to Addendum A1.

[0269] (Addendum A5) Both the first switching element and the second switching element have a first drive electrode and a second drive electrode. Both the first drive electrode and the second drive electrode extend in the arrangement direction. The output via conductor is connected to the end portion closer to the output electrode among the two end portions in the arrangement direction of the second drive electrode of the first switching element. The power supply via conductor is connected to the end portion farther from the output electrode among the two end portions in the arrangement direction of the first drive electrode of the first switching element. The semiconductor device according to Addendum A4.

[0270] (Addendum A6) The output via conductor is connected to the end portion closer to the output electrode among the two end portions in the arrangement direction of the first drive electrode of the second switching element. The ground via conductor is connected to the end portion farther from the output electrode among the two end portions in the arrangement direction of the second drive electrode of the second switching element. The semiconductor device according to Addendum A5.

[0271] (Appendix A7) The specific element is a driver that controls the switching element, In the height direction, the driver and the switching element are arranged at intervals, The semiconductor device includes an element control via conductor that electrically connects the switching element and the driver, The element control via conductor is embedded in the resin layer The semiconductor device according to Appendix A1.

[0272] (Appendix A8) Viewed from the height direction, the driver is arranged at a position overlapping the switching element, The element control via conductor is arranged at a position overlapping both the driver and the switching element when viewed from the height direction, and extends in the height direction The semiconductor device according to Appendix A7.

[0273] (Appendix B1) A first switching element and a second switching element, and A resin layer that seals both the first switching element and the second switching element, where the thickness direction of the resin layer is the height direction, and a semiconductor device comprising: An output electrode formed on the surface of the resin layer, and An output via conductor that electrically connects the first switching element, the second switching element, and the output electrode. A semiconductor device.

[0274] (Appendix B2) The output via conductor is provided in the resin layer The semiconductor device according to Appendix B1.

[0275] (Appendix B3) In the height direction, the output via conductor is provided between the two switching elements and the surface of the resin layer in the resin layer The semiconductor device according to Appendix B2.

[0276] (Appendix B4) The output via conductor includes a first output via conductor connecting the first switching element and the output electrode, and a second output via conductor connecting the second switching element and the output electrode, and in a direction orthogonal to the height direction, the first output via conductor and the second output via conductor are spaced apart from each other. The semiconductor device according to any one of Appendices B1 to B3.

[0277] (Appendix B5) Each of the first output via conductor and the second output via conductor is crank-shaped. The semiconductor device according to Appendix B4.

[0278] (Appendix B6) The output via conductor includes an element connection via provided in the resin layer and penetrating the resin layer in the height direction, and a via connection wiring provided on the resin layer and connecting the element connection via and the output electrode. The semiconductor device according to Appendix B1.

[0279] (Appendix C1) A first switching element and a second switching element, a driver for driving both of the switching elements, and a resin layer for encapsulating each of both of the switching elements and the driver, where the thickness direction of the resin layer is defined as the height direction, and the semiconductor device is such that when two directions intersecting each other in a direction orthogonal to the height direction are defined as a first direction and a second direction, the first switching element and the second switching element are arranged at intervals in the first direction, and both of the switching elements and the driver are arranged at intervals in the second direction. The output electrode formed on the surface of the resin layer, An output via conductor that electrically connects the first switching element, the second switching element, and the output electrode, A semiconductor device comprising an element control via conductor that is embedded in the resin layer and connects the switching element and the driver.

[0280] (Appendix C2) The semiconductor device, A power supply electrode and a ground electrode formed on the surface of the resin layer, A power supply via conductor that penetrates the resin layer in the height direction and electrically connects the first switching element and the power supply electrode, A ground via conductor that penetrates the resin layer in the height direction and electrically connects the second switching element and the ground electrode. The semiconductor device according to Appendix C1.

[0281] (Appendix D1) A step of forming an element encapsulation layer for encapsulating a switching element and a specific element, A step of forming a first through hole in the element encapsulation layer so that the switching element is exposed, A step of embedding an element connection via in the first through hole, A step of forming a via connection wiring on the element encapsulation layer so as to be electrically connected to the element connection via, A step of forming a surface-side resin layer for encapsulating the via connection wiring on the element encapsulation layer, A step of forming a second through hole in the surface-side resin layer so that the via connection wiring is exposed, A step of embedding an electrode connection via in the second through hole, A step of forming an exterior electrode on the surface-side resin layer so as to be connected to the electrode connection via. A method for manufacturing a semiconductor device.

[0282] (Appendix D2) Forming an element encapsulation layer that encapsulates a switching element and a driver that drives the switching element; Forming a first through hole in the element encapsulation layer so that the switching element is exposed; Forming a second through hole in the element encapsulation layer so that the driver is exposed; Embedding an element-side control via in the first through hole; Embedding a driver-side control via in the second through hole; Forming a control connection wiring on the element encapsulation layer so as to electrically connect the element-side control via and the driver-side control via; Forming a surface-side resin layer that encapsulates the control connection wiring on the element encapsulation layer, a method for manufacturing a semiconductor device.

[0283] (Appendix D3) Forming a resin layer that encapsulates a switching element and a specific element; Forming a through hole in the resin layer so that the switching element is exposed; Embedding a via conductor in the through hole; Forming an external electrode on the surface of the resin layer so as to connect to the via conductor, a method for manufacturing a semiconductor device.

[0284] (Appendix D4) The resin layer An element encapsulation layer that encapsulates the switching element and the specific element; And a surface-side resin layer formed on the element encapsulation layer, The through hole penetrates the surface-side resin layer and a portion of the element encapsulation layer that covers the switching element The method for manufacturing a semiconductor device according to Appendix D3. (Appendix E1) a switching element, a specific element, a resin layer that seals the switching element and the specific element, and a semiconductor device having the thickness direction of the resin layer as the height direction, an external electrode formed on the surface of the resin layer and at least partially overlapping the switching element when viewed from the height direction, an element driving via conductor that penetrates the resin layer in the height direction and electrically connects the switching element and the external electrode. A semiconductor device. (Appendix E2) a support layer that supports each of the switching element and the specific element, the resin layer is formed on the support layer, the switching element has an element main surface facing the side opposite to the support layer in the height direction, a first driving electrode, a second driving electrode, and a control electrode are respectively formed on the element main surface, the surface of the resin layer where the external electrode is formed is disposed on the side opposite to the support layer in the resin layer in the height direction The semiconductor device according to Appendix E1. (Appendix E3) the support layer has a first metal layer on which the switching element is mounted, a second metal layer on which the specific element is mounted, and an insulating layer that electrically insulates the first metal layer and the second metal layer The semiconductor device according to Appendix E2. (Appendix E4) the element driving via conductor is disposed at a position overlapping both the external electrode and the switching element when viewed from the height direction and extends in the height direction The semiconductor device according to any one of Appendices E1 to E3. (Appendix E5) the switching element is a first switching element, the specific element is a second switching element connected in series to the first switching element, an output electrode is formed on the surface of the resin layer, the semiconductor device includes an output via conductor provided in the resin layer, The output via conductor electrically connects the output electrode, the first switching element, and the second switching element. The semiconductor device according to Supplementary Note E1. (Supplementary Note E6) The first switching element and the second switching element are arranged at intervals in a direction orthogonal to the height direction. Viewed from the height direction, the output electrode is arranged between the first switching element and the second switching element. The semiconductor device according to Supplementary Note E5. (Supplementary Note E7) The output via conductor is crank-shaped. The semiconductor device according to Supplementary Note E6. (Supplementary Note E8) Both the first switching element and the second switching element have a first drive electrode, a second drive electrode, and a control electrode. The output via conductor includes a first output via conductor connecting the second drive electrode of the first switching element and the output electrode, and a second output via conductor connecting the first drive electrode of the second switching element and the output electrode. The first output via conductor and the second output via conductor are arranged at intervals in the arrangement direction of the first switching element and the second switching element. The semiconductor device according to Supplementary Note E7. (Supplementary Note E9) The resin layer includes an element sealing layer that seals the first switching element and the second switching element, and a surface-side resin layer formed on the element sealing layer. The first output via conductor includes a first element connection via electrically connected to the second drive electrode of the first switching element, a first electrode connection via connected to the output electrode, and a first via connection wiring connecting the first element connection via and the first electrode connection via. The second output via conductor includes a second element connection via electrically connected to the first drive electrode of the second switching element, a second electrode connection via connected to the output electrode, and a second via connection wiring connecting the second element connection via and the second electrode connection via. Each of the first element connection via and the second element connection via penetrates the element sealing layer in the height direction. Each of the first electrode connection via and the second electrode connection via penetrates the surface-side resin layer in the height direction. Each of the first via connection wiring and the second via connection wiring is provided on the element sealing layer. The semiconductor device according to Supplementary Note E8. (Supplementary Note E10) It has a support layer that supports the first switching element and the second switching element, The resin layer is formed on the support layer, The support layer has a metal layer on which the second switching element is mounted, The external electrode is an electrode connected to the ground and has a ground electrode arranged at an interval from the output electrode in a direction orthogonal to the height direction, The element driving via conductor has a ground via conductor that electrically connects the second switching element and the ground electrode, The ground electrode has an overhanging portion that protrudes toward the side opposite to the output electrode with respect to the second switching element when viewed from the height direction and overlaps with the metal layer, The semiconductor device is formed in a portion of the resin layer that overlaps with the overhanging portion when viewed from the height direction, and includes a connection via conductor that penetrates the resin layer in the height direction and connects the overhanging portion and the metal layer, The semiconductor device according to any one of Supplementary Notes E5 to E9. (Supplementary Note E11) Both the first switching element and the second switching element are made of GaN. The semiconductor device according to any one of claims E5 to E10. (Supplementary Note E12) The specific element is a driver that controls the switching element, In a direction orthogonal to the height direction, the driver and the switching element are arranged at an interval, The semiconductor device includes an element control via conductor that electrically connects the switching element and the driver, The element control via conductor is embedded in the resin layer. The semiconductor device according to Supplementary Note E1. (Supplementary Note E13) The resin layer, An element encapsulation layer that encapsulates the switching element, And a surface-side resin layer formed on the element encapsulation layer, The switching element has a control electrode that controls the switching element, The element control via conductor, A driver-side control via electrically connected to the driver, An element-side control via electrically connected to the control electrode, And a control connection wiring that connects the driver-side control via and the element-side control via, The driver-side control via penetrates a portion of the element encapsulation layer that covers the driver in the height direction in the height direction, The element-side control via penetrates in the height direction a portion of the element encapsulation layer that covers the switching element in the height direction. The control connection wiring is provided on the element encapsulation layer. The semiconductor device according to Supplementary Note E12. (Supplementary Note E14) The exterior electrode has a driver connection electrode that is electrically connected to the driver. The semiconductor device includes a via conductor for the driver provided in the resin layer. The via conductor for the driver electrically connects the driver connection electrode and the driver. The semiconductor device according to Supplementary Note E13. (Supplementary Note E15) When viewed from the height direction, the driver connection electrode is disposed at a position that does not overlap with the driver. The via conductor for the driver is in a crank shape. The semiconductor device according to Supplementary Note E14. (Supplementary Note E16) The via conductor for the driver includes a via for driver connection electrically connected to the driver, a via for driver electrode connection connected to the driver connection electrode, and a wiring for via connection that connects the via for driver connection and the via for driver electrode connection. The via for driver connection penetrates the element encapsulation layer in the height direction. The via for driver electrode connection penetrates the surface-side resin layer in the height direction. The wiring for via connection is provided on the element encapsulation layer. The semiconductor device according to Supplementary Note E15. (Supplementary Note E17) A switching element, a driver that drives the switching element, and a resin layer that encapsulates both the switching element and the driver, the semiconductor device having a thickness direction of the resin layer as the height direction, wherein the switching element and the driver are arranged with a gap in a direction orthogonal to the height direction, and an element control via conductor that connects the switching element and the driver is embedded in the resin layer. Semiconductor device.

Description of Symbols

[0285] 10…Semiconductor device 20…Support layer 30…Switching element 30A…First switching element 30B…Second switching element 31…Main surface of the element 31A…First main surface of the element 31B…Second main surface of the element 32A…First inner surface of the element 32B…Second inner surface of the element 31AA, 31BA…First drive pad electrode (First drive electrode) 31AB, 31BB…Second drive pad electrode (Second drive electrode) 31AC, 31BC…Control pad electrode (Control electrode) 33A…First drive pad electrode (First drive electrode) 33B…Second drive pad electrode (Second drive electrode) 33C…Control pad electrode (Control electrode) 40…Driver 43…Driver pad electrode 50…Resin layer 50A…Element encapsulation layer 50B…Surface-side resin layer 51…Main surface of the resin (Surface of the resin layer) 57…Interface 60…Via conductor 61…Power supply via conductor (Element drive via conductor) 62…Ground via conductor (Element drive via conductor) 63…Output via conductor 63P…First output via conductor 63Q…Second output via conductor 63pa…First element connection via 63pb…First electrode connection via 63pc…First via connection wiring 63qa…Second element connection via 63qb…Second electrode connection via 63qc…Second via connection wiring 64…Element control via conductor 64P…First control via conductor 64Q…Second control via conductor 64pa…First element-side control via 64pb…First driver-side control via 64pc…First control connection wiring 64qa…Second element-side control via 64qb…Second driver-side control via 64qc…Second control connection wiring 64a... Element side control via 64b... Driver side control via 64c... Control connection wiring 65... Driver via conductor 65a... Driver connection via 65b... Driver electrode connection via 65c... Via connection wiring 70... Exterior electrode 71... Power supply electrode 72... Ground electrode 73... Output electrode 74... Driver connection electrode 81... Metal layer (first metal layer) 82... Metal layer (second metal layer) 83... Metal layer (second metal layer) 84... Connection via conductor 85... Insulation layer 850A... Element encapsulation layer 850B... Surface side resin layer

Claims

1. A first switching element, a second switching element connected in series with the first switching element, a support layer that supports the first switching element and the second switching element and has a metal layer on which the second switching element is mounted, a resin layer formed on the support layer and sealing the first switching element and the second switching element, an exterior electrode formed on the surface of the resin layer and at least partially overlapping the first switching element when viewed from the height direction, which is the thickness direction of the resin layer, an output electrode formed on the surface of the resin layer, an element driving via conductor that penetrates the resin layer in the height direction and electrically connects the first switching element and the exterior electrode, an output via conductor provided in the resin layer and electrically connecting the output electrode, the first switching element, and the second switching element, A semiconductor device comprising: The exterior electrode is an electrode connected to ground and has a ground electrode disposed at a distance from the output electrode in a direction orthogonal to the height direction, The element driving via conductor has a ground via conductor that electrically connects the second switching element and the ground electrode, The ground electrode has an overhanging portion that protrudes toward the side opposite to the output electrode with respect to the second switching element and overlaps the metal layer when viewed from the height direction, The semiconductor device includes a connection via conductor formed in a portion of the resin layer that overlaps the overhanging portion when viewed from the height direction and penetrating the resin layer in the height direction to connect the overhanging portion and the metal layer. Semiconductor device.

2. The first switching element and the second switching element are disposed at intervals in a direction orthogonal to the height direction, When viewed from the height direction, the output electrode is disposed between the first switching element and the second switching element. The semiconductor device according to Claim 1.

3. The output via conductor is in a crank shape. The semiconductor device according to Claim 2.

4. Both the first switching element and the second switching element have a first drive electrode, a second drive electrode, and a control electrode, The output via conductor is a first output via conductor connecting the second drive electrode of the first switching element and the output electrode; a second output via conductor connecting the first drive electrode of the second switching element and the output electrode; and the first output via conductor and the second output via conductor are arranged at intervals in the arrangement direction of the first switching element and the second switching element. The semiconductor device according to claim 3.

5. The resin layer has an element encapsulation layer that encapsulates the first switching element and the second switching element, and a surface-side resin layer formed on the element encapsulation layer. The first output via conductor has a first element connection via electrically connected to the second drive electrode of the first switching element, a first electrode connection via connected to the output electrode, and a first via connection wiring connecting the first element connection via and the first electrode connection via. The second output via conductor has a second element connection via electrically connected to the first drive electrode of the second switching element, a second electrode connection via connected to the output electrode, and a second via connection wiring connecting the second element connection via and the second electrode connection via. Each of the first element connection via and the second element connection via penetrates the element encapsulation layer in the height direction, each of the first electrode connection via and the second electrode connection via penetrates the surface-side resin layer in the height direction, and each of the first via connection wiring and the second via connection wiring is provided on the element encapsulation layer. The semiconductor device according to claim 4.

6. Both the first switching element and the second switching element are made of GaN. The semiconductor device according to any one of claims 1 to 5.

Images