WO2018173511A1 - Semiconductor device - Google Patents

Abstract

A semiconductor device comprises: a semiconductor chip (12) that has an electrode (121); an electroconductive member (18) that includes a metal substrate (186), and has a mounting part (182) for the semiconductor chip and a surrounding part (183) surrounding the mounting part (182) on the surface opposing the electrode; solder (16) interposed between the electrode and the mounting part, the solder connecting the electrode and the electroconductive member; and a sealing resin body (14) that integrally seals the semiconductor chip, at least the opposing surface of the electroconductive member, and the solder. The electroconductive member has: a bonding part (184) provided as a surrounding part so as to surround the mounting part, the sealing resin body being bonded to the bonding part (184); and a non-bonding part (185) which is provided between the mounting part and the bonding part, the solder not being connected to the non-bonding part (185), and the non-bonding part (185) having lesser bondability to the sealing resin body than does the bonding part.

Description

Semiconductor device Cross-reference of related applications

This application is based on Japanese Patent Application No. 2017-56316 filed on Mar. 22, 2017 and Japanese Patent Application No. 2017-117217 filed on Jun. 14, 2017. The description is incorporated by reference.

The present disclosure relates to a semiconductor device.

Patent Document 1 discloses a semiconductor device in which an electrode of a semiconductor chip and a conductive member are connected via solder and are integrally sealed with a sealing resin body. The conductive member has, on the surface facing the electrode, a mounting portion of the semiconductor chip via solder and a surrounding portion surrounding the mounting portion. And the contact part which the sealing resin body closely_contact | adheres to the surrounding part is provided.

JP2016-197706 A

In the configuration in which the entire peripheral portion is a close contact portion, the close contact portion is adjacent to the mounting portion, that is, the solder. The present inventor has found the following. In the conventional configuration, the thermal stress in the shear direction increases at the interface between the solder-side end of the contact portion and the sealing resin body, so the contact portion is provided to suppress the peeling of the sealing resin body. Regardless, the sealing resin body may be peeled off from the contact portion.

An object of the present disclosure is to provide a semiconductor device that can suppress the peeling of the sealing resin body from the conductive member.

According to one aspect of the present disclosure, a semiconductor device includes a semiconductor chip having an electrode, a metal base, and a mounting portion of the semiconductor chip and a surrounding portion surrounding the mounting portion on a surface facing the electrode. A conductive member, a solder interposed between the electrode and the mounting portion, and connecting the electrode and the conductive member; a semiconductor chip; at least a facing surface of the metal member; and a sealing resin body that integrally seals the solder . A conductive member is provided as a peripheral part so as to surround the mounting part. The conductive member is provided between the close contact portion where the sealing resin body is in close contact with the mounting portion and the close contact portion, solder is not connected, and the close contact with the sealing resin body is lower than the close contact portion. Part.

According to this semiconductor device, the non-contact portion where the solder is not connected and the sealing resin body is not in close contact is provided between the mounting portion and the close contact portion. Thereby, a mounting part, ie, solder, does not adjoin a contact part. Therefore, the thermal stress in the shearing direction acting on the interface between the solder-side end portion and the sealing resin body in the close contact portion can be reduced, and the sealing resin body can be prevented from peeling from the close contact portion.

The above and other objects, features, and advantages of the present disclosure will become more apparent from the following detailed description with reference to the accompanying drawings. In the accompanying drawings,
FIG. 1 is a plan view showing a schematic configuration of the semiconductor device according to the first embodiment. FIG. 2 is a sectional view taken along line II-II in FIG. FIG. 3 is a plan view of the first heat sink as viewed from the facing surface side. 4 is an enlarged cross-sectional view of a region IV indicated by a dashed line in FIG. FIG. 5 is a diagram illustrating the relationship between the width of the non-contact portion and the shear stress acting on the end portion of the close-contact portion, FIG. 6 is a diagram showing the relationship between the width of the non-contact portion and solder strain in each element thickness, FIG. 7 is a diagram showing the relationship between the width of the contact portion and the solder strain, FIG. 8 is a cross-sectional view showing a first modification, corresponding to FIG. FIG. 9 is a sectional view showing a schematic configuration of the semiconductor device according to the second embodiment, corresponding to FIG. FIG. 10 is a cross-sectional view showing a schematic configuration of the semiconductor device according to the third embodiment, corresponding to FIG. FIG. 11 is a diagram illustrating a planar arrangement of the non-contact portion and the second contact portion, FIG. 12 is an enlarged cross-sectional view of a region XII indicated by a dashed line in FIG. FIG. 13 is a cross-sectional view showing a state in which a void is generated on the non-contact portion, corresponding to FIG. FIG. 14 is a diagram showing the relationship between the presence or absence of voids and solder strain, FIG. 15 is a diagram showing a SAT image from the first heat sink side, FIG. 16 is a diagram showing a SAT image from the second heat sink side, FIG. 17 is a diagram illustrating a planar arrangement of the non-contact portion and the second contact portion in the semiconductor device according to the fourth embodiment, and corresponds to FIG. FIG. 18 is a plan view showing a second modification, corresponding to FIG. FIG. 19 is a plan view showing a third modification, corresponding to FIG. FIG. 20 is a plan view showing a fourth modification, and corresponds to FIG.

A plurality of embodiments will be described with reference to the drawings. In several embodiments, functionally and / or structurally corresponding parts are given the same reference numerals. In the following, the thickness direction of the semiconductor substrate is indicated as the Z direction, and one direction orthogonal to the Z direction is indicated as the X direction. A direction perpendicular to both the Z direction and the X direction is referred to as a Y direction. Unless otherwise specified, the shape along the XY plane defined by the X direction and the Y direction is a planar shape.

(First embodiment)
First, a schematic configuration of the semiconductor device will be described with reference to FIGS.

As shown in FIGS. 1 and 2, the semiconductor device 10 includes a semiconductor chip 12, a sealing resin body 14, a first heat sink 18, a main terminal 20, a signal terminal 22, a terminal 26, a second heat sink 30, and a main terminal 32. It has. Such a semiconductor device 10 is applied to, for example, a three-phase inverter constituting an in-vehicle power conversion device. The semiconductor device 10 constitutes one of six arms constituting a three-phase inverter.

The semiconductor chip 12 is formed by forming an element such as an insulated gate bipolar transistor (IGBT) on a semiconductor substrate 120 such as silicon or silicon carbide. In the present embodiment, an n-channel IGBT is formed, and a free-wheeling diode (FWD) connected in reverse parallel to the IGBT is also formed. That is, RC (Reverse Conducting) -IGBT is formed on the semiconductor substrate 120. The semiconductor substrate 120 has a substantially rectangular planar shape.

The IGBT and FWD have a so-called vertical structure so that current flows in the Z direction, which is the thickness direction of the semiconductor substrate 120. Main electrodes are formed on both sides of the semiconductor substrate 120 in the Z direction. As the main electrode, a collector electrode 121 is formed on one surface, and an emitter electrode 122 is formed on the back surface opposite to the one surface. The collector electrode 121 corresponds to an electrode. One surface is also referred to as the front surface or the first surface, and the back surface is also referred to as the second surface.

In the present embodiment, the collector electrode 121 also serves as an FWD cathode electrode, and the emitter electrode 122 also serves as an anode electrode. The collector electrode 121 is formed on the entire back surface of the semiconductor substrate 120. A protective film 123 such as polyimide is formed on one surface of the semiconductor substrate 120, and the emitter electrode 122 is exposed from the protective film 123. That is, the emitter electrode 122 is formed on a part of one surface side of the semiconductor substrate 120. A signal pad (not shown) is also exposed from the protective film 123. The signal pads also include gate electrode pads.

The sealing resin body 14 integrally seals the semiconductor chip 12 and the components of the semiconductor device 10 other than the semiconductor chip 12. The sealing resin body 14 is a resin molded body. The sealing resin body 14 is formed using, for example, an epoxy resin. In this embodiment, the sealing resin body 14 is formed by a transfer molding method.

The sealing resin body 14 has a substantially rectangular planar shape. The sealing resin body 14 has, as a surface in the Z direction, one surface 140 that is one surface and a back surface 141 opposite to the one surface 140. The one surface 140 and the back surface 141 are substantially flat surfaces. Moreover, the sealing resin body 14 has the side surface 142 as a part of surface. The side surface 142 is continuous with the one surface 140 and the back surface 141.

The first heat sink 18 is connected to the collector electrode 121 of the semiconductor chip 12 via the solder 16. The first heat sink 18 corresponds to a conductive member. The first heat sink 18 radiates heat generated by the semiconductor chip 12 to the outside of the semiconductor device 10. The first heat sink 18 electrically relays the semiconductor chip 12 and a main terminal 20 described later.

The first heat sink 18 has a facing surface 180 facing the collector electrode 121 and a back surface 181 opposite to the facing surface 180 as the surface in the Z direction. As shown in FIGS. 2 and 3, the first heat sink 18 has a mounting portion 182 and a peripheral portion 183 on the facing surface 180. The mounting portion 182 is a portion of the facing surface 180 where the solder 16 is connected, that is, a portion where the semiconductor chip 12 is mounted. The mounting portion 182 includes at least a portion that overlaps with the collector electrode 121 (semiconductor chip 12) in plan view from the Z direction. The peripheral portion 183 is a portion excluding the mounting portion 182 on the facing surface 180. The peripheral portion 183 surrounds the mounting portion 182.

The peripheral portion 183 includes a close contact portion 184 to which the sealing resin body 14 is in close contact and a non-contact portion 185 to which the sealing resin body 14 is not in close contact. The contact portion 184 surrounds the mounting portion 182 at a position away from the mounting portion 182 so as not to be adjacent to the mounting portion 182. A portion between the mounting portion 182 and the contact portion 184 is a non-contact portion 185. The non-contact portion 185 surrounds the mounting portion 182 adjacent to the mounting portion 182. In this embodiment, the close contact part 184 is formed by the roughening process mentioned later. That is, the contact portion 184 is configured by the roughened portion. Further, the entire portion from the non-contact portion 185 to the outer peripheral edge of the facing surface 180 is a contact portion 184. The width of the non-contact portion 185 is substantially constant over the entire circumference.

The non-contact part 185 has lower adhesion to the sealing resin body 14 than the contact part 184. As a result, the sealing resin body 14 does not adhere to the non-contact portion 185 but closely contacts only the contact portion 184. Thus, the sealing resin body 14 is intentionally peeled off at the non-contact portion 185. Details of the first heat sink 18 including the peripheral portion 183 will be described later.

The back surface 181 of the first heat sink 18 is exposed from the sealing resin body 14. The back surface 181 is substantially flush with the one surface 140. Thus, the back surface 181 is a heat radiating surface that radiates heat to the outside of the semiconductor device 10. The surface excluding the back surface 181, that is, the facing surface 180 is also covered with the sealing resin body 14.

The main terminal 20 is connected to the first heat sink 18. The main terminal 20 is electrically connected to the collector electrode 121 via the first heat sink 18. The main terminal 20 extends from the first heat sink 18 in the Y direction, and protrudes from one of the side surfaces 142 of the sealing resin body 14 to the outside. The main terminal 20 may be formed integrally with the first heat sink 18 as a part of the lead frame, or the main terminal 20 of another member may be connected to the first heat sink 18. In the present embodiment, the main terminal 20 is formed integrally with the first heat sink 18. The main terminal 20 is thinner than the first heat sink 18. The main terminal 20 is substantially flush with the facing surface 180 of the first heat sink 18.

The signal terminal 22 is electrically connected to the pad of the semiconductor chip 12 via a bonding wire (not shown). As shown in FIG. 1, the signal terminal 22 extends in the Y direction. The signal terminal 22 protrudes to the outside from a surface opposite to the side surface 142 from which the main terminal 20 protrudes.

A terminal 26 is connected to the emitter electrode 122 of the semiconductor chip 12 via a solder 24. The terminal 26 is interposed between the semiconductor chip 12 and the second heat sink 30. The terminal 26 has a spacer function for ensuring the height of the bonding wire described above. For this reason, it is not always necessary. For example, a convex portion may be provided on the second heat sink 30 and the convex portion may have a spacer function.

The terminal 26 includes a metal substrate such as Cu. The terminal 26 electrically relays the emitter electrode 122 of the semiconductor chip 12 and the second heat sink 30. The heat generated by the semiconductor chip 12 is transferred to the second heat sink 30 via the terminal 26.

A second heat sink 30 is connected to the surface of the terminal 26 opposite to the semiconductor chip 12 via a solder 28. The second heat sink 30 radiates heat generated by the semiconductor chip 12 to the outside of the semiconductor device 10. The second heat sink 30 electrically relays the semiconductor chip 12 and a main terminal 32 described later. The second heat sink 30 includes a genus base material such as Cu.

The second heat sink 30 has a facing surface 300 facing the terminal 26 and a back surface 301 opposite to the facing surface 300 as the surface in the Z direction. The second heat sink 30 has a groove portion 302 for accommodating the overflowing solder 28 on the facing surface 300. By accommodating the solder 28 overflowing in the groove portion 302, it is possible to suppress the solder 28 from spreading to the semiconductor chip 12 side through the side surface of the terminal 26.

The back surface 301 of the second heat sink 30 is exposed from the sealing resin body 14. The back surface 301 is exposed to be substantially flush with the back surface 141. Thus, the back surface 301 is a heat radiating surface that radiates heat to the outside of the semiconductor device 10. The surface excluding the back surface 301, that is, the facing surface 300 is also covered with the sealing resin body 14.

The main terminal 32 is connected to the second heat sink 30. The main terminal 32 is electrically connected to the emitter electrode 122 via the second heat sink 30. The main terminal 32 extends from the second heat sink 30 in the Y direction and on the same side as the main terminal 20. The main terminal 32 protrudes from the same side surface 142 as the main terminal 20. The main terminal 32 may be formed integrally with the second heat sink 30 as a part of the lead frame, or a separate main terminal 32 may be connected to the second heat sink 30. In the present embodiment, the main terminal 32 is formed integrally with the second heat sink 30. The main terminal 32 is thinner than the second heat sink 30. The main terminal 32 is substantially flush with the facing surface 300 of the second heat sink 30.

In the semiconductor device 10 configured as described above, the sealing resin body 14 causes the semiconductor chip 12, a part of the first heat sink 18, a part of each of the main terminals 20 and 32, a part of the signal terminal 22, and a terminal 26. And a part of the second heat sink 30 are integrally sealed. In the semiconductor device 10, the semiconductor chip 12 constituting one arm is sealed with the sealing resin body 14. For this reason, the semiconductor device 10 is also referred to as a 1 in 1 package.

The first heat sink 18 and the second heat sink 30 are cut together with the sealing resin body 14. Therefore, the one surface 140 and the back surface 181 are cutting surfaces and are substantially flush with each other. Similarly, the back surface 141 and the back surface 301 are cutting surfaces and are substantially flush with each other. As described above, the semiconductor device 10 has a double-sided heat dissipation structure in which the back surfaces 181 and 301 are both exposed from the sealing resin body 14.

In addition, as the solder 16, 24, 28, fluxless solder is used. The one surface 140 and the back surface 181 are not limited to cutting surfaces. The back surface 141 and the back surface 301 are not limited to the cutting surface. The back surface 181 and 301 may be exposed from the sealing resin body 14 without being cut by bringing the back surface 181 and 301 into contact with the wall surface of the molding die of the sealing resin body 14.

Next, the detailed structure of the first heat sink 18 will be described with reference to FIG.

As shown in FIG. 4, the first heat sink 18 includes a base 186 formed using a metal material such as Cu, and a coating 187 provided on at least the facing surface 180 side of the surface of the base 186. Have. The base material 186 corresponds to a metal base material. The base material 186 has a substantially rectangular parallelepiped shape. The film 187 is a metal thin film 188 formed on the surface of the base material 186 and an oxide of the same metal as the main component metal constituting the metal thin film 188, and the uneven oxide film 189 whose surface continuously forms unevenness. have.

In this embodiment, the metal thin film 188 is mainly composed of Ni. The metal thin film 188 is formed by, for example, plating or vapor deposition. The metal thin film 188 is formed on the surface of the substrate 186 by, for example, electroless Ni plating. The metal thin film 188 contains P (phosphorus) in addition to Ni as the main component.

The metal thin film 188 is formed on the surface of the base material 186 except for the back surface 181 side. On the facing surface 180 side, a plurality of recesses 188a are formed in a portion of the surface of the metal thin film 188 corresponding to the contact portion 184. That is, the mounting portion 182 and the non-contact portion 185 are not formed with the recess 188a. In the portion where the recess 188a is not formed, the thickness of the metal thin film 188 is, for example, about 10 μm. In other words, the film thickness before laser light irradiation described later is about 10 μm.

The recess 188a is formed by irradiation with pulsed laser light. One recess 188a is formed for each pulse. Adjacent recesses 188a are continuous in the laser beam scanning direction. The plurality of recesses 188a are continuous in the X direction and also continuous in the Y direction. In the contact portion 184, the surface of the metal thin film 188 has a scale shape due to the plurality of recesses 188a. A portion corresponding to the contact portion 184 is a laser light irradiation area, and a portion corresponding to the mounting portion 182 and the non-contact portion 185 is a non-irradiation area.

The width of each recess 188a is set to 5 μm to 300 μm. The depth of the recess 188a is 0.5 μm to 5 μm. If the depth of the recess 188a is less than 0.5 μm, the surface of the metal thin film 188 is not sufficiently melted and deposited by laser light irradiation, and the uneven oxide film 189 described later is difficult to form. If the depth of the recess 188a is deeper than 5 μm, the surface of the metal thin film 188 is likely to be melted and scattered, and surface formation by melting and scattering becomes more dominant than vapor deposition, and the uneven oxide film 189 becomes difficult to form.

The uneven oxide film 189 is formed on the metal thin film 188 on the facing surface 180 side. The uneven oxide film 189 is not formed in the mounting portion 182 but is formed in the peripheral portion 183, that is, the close contact portion 184 and the non-contact portion 185. The uneven oxide film 189 is formed by oxidizing the metal constituting the metal thin film 188 by irradiating the metal thin film 188 with laser light. The uneven oxide film 189 is an oxide film formed on the surface of the metal thin film 188 by oxidizing the surface layer of the metal thin film 188. Since the uneven oxide film 189 is formed by laser light irradiation, it can be said to be a laser irradiation film.

In the present embodiment, among the components constituting the uneven oxide film 189, 80% is NI ₂ O ₃ , 10% is NiO, and 10% is Ni. Thus, the main component of the uneven oxide film 189 is an oxide of Ni which is the main component of the metal thin film 188.

The average film thickness of the uneven oxide film 189 is 10 nm to several hundreds nm in the close contact portion 184, that is, the laser light irradiation area. The uneven oxide film 189 is formed following the unevenness of the surface of the metal thin film 188 having the recess 188a. Further, unevenness is formed on the surface of the uneven oxide film 189 at a pitch finer than the width of the recess 188a. That is, very fine unevenness (roughened part) is formed. In other words, the plurality of convex portions 189a (columnar bodies) are formed at a fine pitch. For example, the average width of the protrusions 189a is 1 nm to 300 nm, and the average interval between the protrusions 189a is 1 nm to 300 nm. Further, the average height of the convex portion 189a is set to 10 nm to several hundred nm.

Thus, the contact portion 184 is constituted by the uneven oxide film 189 having very fine unevenness formed on the surface. The sealing resin body 14 is entangled with the convex portion 189a on the surface of the uneven oxide film 189, and an anchor effect is generated. Moreover, since the height of the convex part 189a is higher than the non-contact part 185, a contact area with the sealing resin body 14 increases. Therefore, the sealing resin body 14 is in close contact with the close contact portion 184 of the facing surface 180.

The concavo-convex oxide film 189 is formed by irradiating the metal thin film 188 with laser light and melting and vapor-depositing the surface of the metal thin film 188. Also formed. In the present embodiment, the uneven oxide film 189 is formed on the entire area of the non-contact portion 185 in the non-irradiated area of the laser beam, and the uneven oxide film 189 is not formed on the mounting portion 182. The width W1 (see FIG. 3) of the non-contact portion 185 having the uneven oxide film 189 over the entire area is set to 0.2 mm to 0.3 mm.

However, since the laser beam is not directly irradiated, the average film thickness of the uneven oxide film 189 in the non-contact portion 185 is smaller than the average film thickness of the uneven oxide film 189 in the contact portion 184 and is naturally It is thicker than the oxide film. Specifically, it is set to 0.1 nm to 10 nm. Further, the height of the convex portion 189 a on the surface of the uneven oxide film 189 is also made lower than the close contact portion 184. Specifically, it is set to 0.1 nm to 10 nm. Note that the average width and average interval of the convex portions 189a are approximately the same as those of the contact portions 184.

By having the uneven oxide film 189 described above, the adhesion of the non-adhered portion 185 to the sealing resin body 14 is made lower than that of the adhered portion 184. As a result, the sealing resin body 14 does not adhere to the non-contact portion 185. Thus, the sealing resin body 14 is intentionally peeled off at the non-contact portion 185. Further, by having the uneven oxide film 189, the wettability of the non-contact portion 185 with respect to the solder 16 is made lower than that of the mounting portion 182. That is, the solder 16 is difficult to wet and spread from the mounting portion 182 to the non-contact portion 185 side. As a result, the solder 16 is not connected to the non-contact portion 185.

When forming the semiconductor device 10 described above, an uneven oxide film 189 is formed on the first heat sink 18 in advance before reflowing the solders 16, 24, 28. In forming the uneven oxide film 189, a pulse oscillation laser beam is irradiated to the formation region of the contact portion 184 in the surface of the metal thin film 188 on the facing surface 180 side of the first heat sink 18. Laser light is scanned in the X direction so that adjacent laser light spots (irradiation range by one pulse) partially overlap in the X direction. Further, the laser beam is scanned in the Y direction so that adjacent laser beam spots partially overlap in the Y direction. Thereby, the entire region where the contact portion 184 is formed is irradiated with laser light.

The laser light irradiation melts and vaporizes the surface of the metal thin film 188, and a plurality of recesses 188a are formed. In addition, the molten and vaporized metal thin film 188 is deposited on the portion irradiated with the laser light (that is, the formation region of the contact portion 184) and the peripheral portion thereof (that is, the formation region of the non-contact portion 185). As a result, a concavo-convex oxide film 189 having a large film thickness in the close contact portion 184 and a thin film thickness in the non-contact portion 185 is formed. In addition, an uneven oxide film 189 in which the height of the convex portion 189a is high in the close contact portion 184 and the height of the convex portion 189a is low in the non-contact portion 185 is formed.

In the mounting portion 182, the uneven oxide film 189 is not formed on the surface of the metal thin film 188, and a natural oxide film (not shown) is formed. Since this natural oxide film is thinner than the uneven oxide film 189 of the non-contact portion 185, it is reduced and removed during reflow of the solder 16, for example, reduced pressure reflow in a hydrogen atmosphere.

A known method can be applied to the reflow of the solder 16, 24, 28. If necessary, the description in JP-A-2016-197706 can be incorporated.

Next, an example of the effect of the semiconductor device 10 will be described.

The inventors of the present application found the following. When the close contact portion is adjacent to the mounting portion on the facing surface of the first heat sink, that is, when the close contact portion is adjacent to the solder, the thermal stress in the shearing direction causes the interface between the solder end and the sealing resin body in the close contact portion. There is a possibility that the sealing resin body may be peeled from the contact portion even though the contact portion is provided to suppress the peeling of the sealing resin body.

On the other hand, in the semiconductor device 10 of the present embodiment, on the facing surface 180 of the first heat sink 18, between the mounting portion 182 to which the solder 16 is connected and the contact portion 184 to which the sealing resin body 14 is in close contact, A non-contact portion 185 where the solder 16 is not connected and the sealing resin body 14 is not in close contact is provided. By providing the non-contact part 185, the mounting part 182, that is, the solder 16 is not adjacent to the contact part 184. Therefore, the thermal stress in the shearing direction acting on the interface between the end portion on the solder 16 side of the close contact portion 184 and the sealing resin body 14 can be reduced. That is, it is possible to suppress the sealing resin body 14 from being peeled from the close contact portion 184.

FIG. 5 shows the width W1 of the non-contact portion 185 and the thermal stress in the shear direction (hereinafter referred to as shear stress) acting on the interface between the end portion on the solder 16 side of the close contact portion 184 and the sealing resin body 14. The result of the simulation is shown for the relationship. In this simulation, the thickness of the semiconductor substrate 120, that is, the element was 105 μm, the element size was 13.4 mm × 15 mm, the thickness of the solder 16 was 40 μm, and the thickness of the solder 24 was 150 μm. Further, the temperature of the sealing resin body 14 was changed from 180 ° C. to −40 ° C., and the shear stress at −40 ° C. was obtained.

As shown in FIG. 5, the shear stress when the width W1 of the non-contact portion 185 is 0 mm, that is, when the close contact portion 184 is adjacent to the mounting portion 182 is 40.5 MPa, and the shear stress when the width W1 is 0.1 mm. Was 22.3 MPa. The shear stress when the width W1 was 0.2 mm was 18.5 MPa, the shear stress when the width W1 was 0.5 mm was 10.0 MPa, and the shear stress when the width W1 was 1.0 mm was 5.1 MPa. Furthermore, the shear stress when the width W1 was 1.5 mm was 2.6 MPa, and the shear stress when the width W1 was 1.95 mm was 0.8 MPa.

Thus, also from the simulation results, it is clear that the shear stress can be reduced by providing the non-contact portion 185 rather than the configuration in which the close contact portion 184 is adjacent to the mounting portion 182.

In particular, in this embodiment, the width W1 of the non-contact portion 185 is set to 0.2 mm to 0.3 mm. As shown in FIG. 5, when the width W1 is 0.2 mm or more, the shear stress can be suppressed to 20 MPa or less. When the shear stress is set to 20 MPa or less, it has been confirmed that the sealing resin body 14 can be effectively prevented from being peeled off from the close contact portion 184 in the above-described thermal cycle test. Therefore, in this embodiment, since the width W1 is set to 0.2 mm or more, it is possible to effectively suppress the sealing resin body 14 from being peeled off from the contact portion 184.

FIG. 6 shows the result of simulation regarding the relationship between the width W1 of the non-contact portion 185 and the distortion (plastic strain) of the solder 16 due to thermal stress. In this simulation, the element size was 13.4 mm × 15 mm, the thickness of the solder 16 was 40 μm, and the thickness of the solder 24 was 40 μm. As shown in FIG. 6, the element thickness was set at three levels of 0.08 mm (80 μm), 0.12 mm, and 0.18 mm. Further, as shown in FIG. 6, regarding the element thickness, the circle mark corresponds to 0.08 mm, the square mark corresponds to 0.12 mm, and the triangle mark corresponds to 0.18 mm. Then, the temperature of the sealing resin body 14 was changed from 180 ° C. to −40 ° C., and the solder strain (%) was obtained from the displacement of the solder 16 at −40 ° C. with respect to 180 ° C.

FIG. 6 clearly shows that the solder strain can be reduced at any element thickness when the width W1 is 0.5 mm or less. In the present embodiment, since the width W1 is set to 0.2 mm to 0.3 mm, solder strain, that is, thermal stress acting on the solder 16 can be reduced regardless of the element thickness. That is, the connection reliability of the solder 16 can be improved regardless of the element thickness. According to the present embodiment, it is possible to improve the connection reliability of the solder 16 while effectively suppressing the sealing resin body 14 from being peeled from the close contact portion 184.

Furthermore, in this embodiment, the uneven oxide film 189 having fine unevenness is formed on the facing surface 180 side of the first heat sink 18 by irradiation with laser light. Then, the portion irradiated with the laser light and having a high height of the convex portion 189 a, that is, a portion where the thickness of the uneven oxide film 189 is thick is defined as the adhesion portion 184. Thus, since the roughening part by laser irradiation is made into the contact part 184, it is easy to form the contact part 184 locally.

Further, a portion around the contact portion 184 and a portion where the height of the convex portion 189a is low, that is, a portion where the thickness of the uneven oxide film 189 is thin is defined as a non-contact portion 185. In this manner, the uneven oxide film 189 is formed over the entire region where the non-contact portion 185 is formed. Fine unevenness is formed on the surface of the uneven oxide film 189, and the wettability of the non-contact portion 185 to the solder 16 is lower than that of the mounting portion 182. Therefore, the area where the solder 16 spreads out can be limited to the inside of the uneven oxide film 189, that is, the inside of the non-contact portion 185. For this reason, it is easy to define the mounting portion 182. In other words, it is easy to obtain the non-contact portion 185 having the desired width W1.

FIG. 7 shows a simulation of the relationship between the minimum width W2 of the contact portion 184 (hereinafter simply referred to as the width W2 of the contact portion 184) and the distortion of the solder 16 (plastic strain) due to thermal stress. Results are shown. In this simulation, the thickness of the element was 105 μm, the size of the element was 13.4 mm × 15 mm, the thickness of the solder 16 was 40 μm, the thickness of the solder 24 was 150 μm, and the width W1 of the non-contact portion 185 was 0.65 mm. Further, the temperature change of the sealing resin body 14 from 180 ° C. to −40 ° C. was performed for 3 cycles, and the solder strain (%) was obtained. FIG. 7 clearly shows that the solder strain exhibits a substantially constant value even when the width W2 is changed to 0.5 mm, 1.0 mm, 1.5 mm, and 2.0 mm.

In the present embodiment, the example in which the uneven oxide film 189 having the protrusion 189a having a low height is formed in the entire area of the non-contact portion 185 is shown. However, as in the first modification shown in FIG. 8, the uneven oxide film 189 having the protrusions 189 a having a low height may be formed only on a part of the non-contact portion 185. In FIG. 8, the width W1 of the non-contact portion 185 is, for example, 1 mm. In the width direction of the non-contact portion 185, a concavo-convex oxide film 189 having a convex portion 189a having a low height is formed on a portion on the close contact portion 184 side, and natural oxidation is performed on the remaining portion, that is, a portion on the mounting portion 182 side. A film 190 is formed.

(Second Embodiment)
This embodiment can refer to the preceding embodiment. For this reason, the description of the parts common to the semiconductor device 10 shown in the previous embodiment is omitted.

As shown in FIG. 9, in the present embodiment, a resin layer 191 that increases the adhesion with the sealing resin body 14 is formed in the formation region of the adhesion portion 184, whereby the sealing resin body 14 is attached to the adhesion portion 184. It is in close contact with. The resin layer 191 is formed on the metal thin film 188. The resin layer 191 is formed only on the close contact portion 184 in the peripheral portion 183 and is not formed on the non-contact portion 185. As a constituent material of the resin layer 191, polyamide, polyimide, polyamideimide, or the like can be used. The resin layer 191 is formed by dispensing or the like. In the non-contact portion 185, a natural oxide film is formed on the metal thin film 188, but the illustration is omitted.

As described above, the non-contact portion 185 is intentionally provided between the mounting portion 182 and the close contact portion 184 by the selective arrangement of the resin layer 191. Thereby, the mounting portion 182 is not adjacent to the contact portion 184. Therefore, the thermal stress in the shearing direction acting on the interface between the end portion on the solder 16 side of the close contact portion 184 and the sealing resin body 14 can be reduced.

(Third embodiment)
This embodiment can refer to the preceding embodiment. For this reason, the description of the parts common to the semiconductor device 10 shown in the previous embodiment is omitted.

As shown in FIGS. 10 and 11, the second heat sink 30 of the present embodiment has a close contact portion 304 in the peripheral portion 303 on the facing surface 300. The close contact portion 304 is a portion of the facing surface 300 where the sealing resin body 14 is in close contact, like the close contact portion 184 provided in the first heat sink 18. FIG. 11 omits a part of the semiconductor device 10, the sealing resin body 14, the signal terminal 22, and the like. In FIG. 11, in order to clarify the overlap between the non-contact portion 185 and the close contact portion 304 in a plan view in the Z direction, the close contact portion 304 is hatched.

Of the facing surface 300 of the second heat sink 30, a portion outside the above-described groove portion 302 is a peripheral portion 303. In a plan view in the Z direction, a portion inside the outer peripheral end of the groove portion 302 is a mounting portion to which the solder 28 is connected. In the present embodiment, the entire area of the peripheral portion 303 is the close contact portion 304. The contact portion 304 overlaps the entire area of the non-contact portion 185 in a plan view in the Z direction. That is, the entire region of the groove portion 302 overlaps with the mounting portion 182 inside the peripheral portion 183.

The close contact portion 304 is formed by the same roughening treatment as the close contact portion 184. That is, the contact portion 304 is configured by the roughening portion. The configuration of the first heat sink 18 is the same as that of the first embodiment.

As shown in FIG. 12, the second heat sink 30 is similar to the first heat sink 18, and the base material 305 formed using a metal material such as Cu and the surface of the base material 305 at least on the facing surface 300 side. It has a coating 306 provided. The base material 305 has a substantially rectangular parallelepiped shape. The film 306 is a metal thin film 307 formed on the surface of the substrate 305 and an oxide of the same metal as the main component metal constituting the metal thin film 307, and the uneven oxide film 308 whose surface continuously forms unevenness. have.

In this embodiment, the portion of the metal thin film 307 where the uneven oxide film 308 is not formed has an electroless Ni plating film and an Au plating film formed on the electroless Ni plating film. On the other hand, the portion where the uneven oxide film 308 is formed has an electroless Ni plating film. As described above, the Au plating film does not exist in the region where the uneven oxide film 308 is formed on the facing surface 300. The reason is that the Au plated film is removed by laser light irradiation and the uneven oxide film 308 is removed from the lower electroless Ni plating film. It is for forming. The metal thin film 307 is formed on the surface of the base material 305 except for the back surface 301 side.

The uneven oxide film 308 is formed in the same manner as the uneven oxide film 189. The uneven oxide film 308 is formed by laser light irradiation. Specifically, a second heat sink 30 on which an electroless Ni plating film and an Au plating film on the electroless Ni plating film are formed is prepared. Then, pulsed laser light is irradiated under the same conditions as the uneven oxide film 189. As a result, while removing the upper Au plating film, the surface layer portion of the lower electroless Ni plating film is melted and vaporized to form the uneven oxide film 308 on the surface of the metal thin film 307.

Since the uneven oxide film 308 is formed by the pulsed laser beam in this way, a plurality of recesses 307 a are formed on the surface of the metal thin film 307 corresponding to the adhesion portion 304, similar to the recesses 188 a of the metal thin film 188. Is done. One recess 307a is formed for each pulse. In addition, a convex portion 308a is formed on the surface of the uneven oxide film 308 in the same manner as the convex portion 189a.

In the present embodiment, the entire area of the peripheral portion 303 of the facing surface 300 is a laser light irradiation area, and the inner side of the outer peripheral edge of the groove 302, that is, the entire mounting portion is a non-irradiation area of the laser light. Therefore, like the non-contact portion 185 adjacent to the close contact portion 184, the uneven oxide film 308 is also formed on the portion adjacent to the close contact portion 304 (for example, the wall surface of the groove portion 302). However, the second heat sink 30 has an Au plating film as the metal thin film 307. Therefore, the solder 28 spreads in the groove 302 due to the effect of Au.

In the present embodiment, the second heat sink 30 corresponds to the second conductive member, and the first heat sink 18 corresponds to the first conductive member. Further, the close contact portion 184 corresponds to a first close contact portion, and the close contact portion 304 corresponds to a second close contact portion. The collector electrode 121 corresponds to the first electrode, and the emitter electrode 122 corresponds to the second electrode.

Next, an example of the effect of the semiconductor device 10 will be described.

As shown in FIG. 13, due to air entrainment at the time of molding the sealing resin body 14, air in the resin tablet that is the material of the sealing resin body 14, vaporization of moisture absorbed, and the like, as shown in FIG. There is a possibility that a void 40 may be formed adjacent to.

FIG. 14 shows the result of simulation regarding the relationship between the presence or absence of the void 40 and the strain (plastic strain) of the solder 16. In this simulation, the element size was 13.4 mm × 15 mm, the thickness of the solder 16 was 40 μm, and the thickness of the solder 24 was 150 μm. The element thickness was 0.08 mm (80 μm). Then, the temperature of the sealing resin body 14 was changed from 180 ° C. to −40 ° C., and the solder strain (%) was obtained from the displacement of the solder 16 at −40 ° C. with respect to 180 ° C.

Note that only peeling in FIG. 14 indicates a case where the sealing resin body 14 is not in close contact with the non-contact portion 185 and the sealing resin body 14 is peeled off from the non-contact portion 185. That is, the case where the void 40 does not arise is shown. On the other hand, the void 1 and the void 2 show the case where the void 40 is generated. The void 1 shows a case where the thickness of the void 40 is equal to the sum of the thickness of the solder 16 and ½ of the thickness of the semiconductor chip 12. The void 2 shows a case where the thickness of the void 40 is equal to the sum of the thickness of the solder 16 and the thickness of the semiconductor chip 12.

As shown in FIG. 14, when the void 40 is generated, it is clear that the solder distortion increases. In addition, it is clear that the thicker (larger) the void 40 is, the larger the solder strain becomes. Thus, when the solder strain increases, the connection reliability of the solder 16 is lowered. For this reason, it is important to detect the presence or absence of the void 40.

An ultrasonic flaw detector (SAT: Scanning Acoustic Tomography) can be used for detection of the void 40. However, even if an attempt is made to detect the void 40 from the first heat sink 18 side in the Z direction, the peeling of the sealing resin body 14 at the non-contact portion 185 exists before the void 40, so the SAT image is shown in FIG. It becomes like this. Reference numeral 41 shown in FIG. 15 indicates a peeling portion of the sealing resin body 14 with respect to the non-contact portion 185. Thus, the void 40 is hidden behind the peeling part 41, and the void 40 and the peeling part 41 cannot be identified. That is, the void 40 cannot be detected.

As shown in the preceding embodiment, in the configuration in which the close contact portion 304 is not provided on the facing surface 300 of the second heat sink 30, the sealing resin body 14 peels from the peripheral portion 303 of the facing surface 300. For this reason, even if it is going to detect the void 40 from the 2nd heat sink 30 side, the peeling part by the side of the 2nd heat sink 30 exists in front of the void 40. FIG. Therefore, the void 40 is hidden behind the peeling portion, and the void 40 cannot be detected.

On the other hand, in this embodiment, the close contact portion 304 is provided on the facing surface 300 of the second heat sink 30, and the close contact portion 304 overlaps the non-contact portion 185 in a plan view in the Z direction. Since the sealing resin body 14 is in close contact with the second heat sink 30 immediately above the non-contact portion 185, the state of the interface between the second heat sink 30 and the sealing resin body 14 does not hinder the detection of the void 40. .

Therefore, by adjusting the depth of the ultrasonic flaw detection, the void 40 positioned in front of the peeling portion 41 can be detected with high accuracy. The SAT image from the second heat sink 30 side is as shown in FIG. As a result, the connection reliability of the solder 16 can be improved, and as a result, a decrease in product life can be suppressed.

Particularly in this embodiment, the close contact portion 304 overlaps the entire non-contact portion 185 in a plan view in the Z direction. Therefore, the void 40 can be detected regardless of where the void 40 occurs in the non-contact portion 185.

In this embodiment, the close contact portion 304 is constituted by the uneven oxide film 308 having very fine unevenness formed on the surface. Accordingly, the sealing resin body 14 can be brought into close contact due to an anchor effect or an increase in contact area. In addition, since the contact portions 184 and 304 can be formed together by laser light irradiation, the manufacturing process can be simplified.

In addition, although the example which makes the whole region of the surrounding part 303 the contact | adherence part 304 was shown, it is not limited to this. The close contact portion 304 may be provided only on a part of the peripheral portion 303 so that the close contact portion 304 overlaps the entire area of the non-adhesive portion 185.

Although the example of the roughening part which has the uneven | corrugated oxide film 189 was shown as the contact | adherence part 184 of the 1st heat sink 18, it is not limited to this. Combination with the adhesion part 184 having the resin layer 191 shown in the second embodiment is also possible.

(Fourth embodiment)
This embodiment can refer to the preceding embodiment. For this reason, the description of the parts common to the semiconductor device 10 shown in the previous embodiment is omitted.

In this embodiment, the close contact portion 304 of the second heat sink 30 is provided so as to overlap only a part of the non-contact portion 185 in a plan view in the Z direction. The other configuration is the same as that of the third embodiment.

In FIG. 17, the contact portion 304 is provided so as to overlap only a part of the circumferential direction in the non-contact portion 185 having a planar rectangular ring shape. Specifically, in the non-contact portion 185, the close contact portion 304 is provided corresponding to a portion where the void 40 is likely to occur due to a gate position of a mold (not shown) for molding the sealing resin body 14. In FIG. 17, a side gate is used, and the contact portion 304 is provided so that the position from the gate includes the farthest position around the semiconductor chip 20 in a plan view in the Z direction. The contact portion 304 is provided corresponding to one of the four corners of the non-contact portion 185.

Thus, even if the contact portion 304 is provided corresponding to only a part of the non-contact portion 185, the void 40 on the non-contact portion 185 can be detected. In the Z direction, the void 40 immediately below the contact portion 304 can be detected.

Note that the position of the contact portion 304 is not limited to the example shown in FIG. For example, when a center gate is employed, the contact portion 304 may be provided near the center of the side of the non-contact portion 185 having a rectangular ring shape as in the second modification shown in FIG.

Further, as in the third modification shown in FIG. 19, the contact portions 304 may be provided at the four corners of the non-contact portion 185 having a rectangular ring shape. According to this, it can be detected whether the void 40 is generated in any of the four corners of the non-contact portion 185.

Further, as in the fourth modified example shown in FIG. 20, the contact portion 304 may be provided so as to overlap only a part of the non-contact portion 185 in the width direction, not in the circumferential direction. This also makes it possible to detect the void 40 immediately below the contact portion 304 in the Z direction.

FIG. 17 to FIG. 20 correspond to FIG. 17 to 20, as in FIG. 11, in order to clarify the overlap between the non-contact portion 185 and the close contact portion 304, the close contact portion 304 is hatched.

The disclosure of this specification is not limited to the illustrated embodiments. The disclosure encompasses the illustrated embodiments and variations by those skilled in the art based thereon. For example, the disclosure is not limited to the combination of elements shown in the embodiments. The disclosure can be implemented in various combinations. The technical scope disclosed is not limited to the description of the embodiments.

Although the example of the 1 in 1 package structure provided with one semiconductor chip 12 was shown as the semiconductor device 10, it is not limited to this. The present invention can also be applied to a configuration including two semiconductor chips 12 constituting the upper and lower arms for one phase and a configuration including six semiconductor chips 12 constituting the upper and lower arms for three phases. That is, the present invention can be applied to a 2in1 package structure and a 6in1 package structure.

Although an example in which the IGBT and the FWD are formed on the same semiconductor chip 12 is shown, the present invention is not limited to this. The IGBT and FWD may be separate chips.

Although the example in which the back surface 181 of the first heat sink 18 and the back surface 301 of the second heat sink 30 are exposed from the sealing resin body 14 is shown, the present invention is not limited thereto. A configuration in which at least one of the rear surfaces 181 and 301 is covered with the sealing resin body 14 may be adopted.

The metal constituting the metal thin film 188 is not limited to Ni. That is, the uneven oxide film 189 is not limited to Ni oxide. The uneven oxide film 189 may be an oxide of the same metal as that constituting the metal thin film 188.

Although the example of the roughening part by laser irradiation was shown as a roughening part which comprises the contact | adherence part 184, it is not limited to this. For example, the sealing resin body 14 may be in close contact with only the close contact portion 184 and the sealing resin body 14 may not be in close contact with the non-contact portion 185 by roughening plating.

Although the semiconductor device 10 includes the semiconductor chip 12, the sealing resin body 14, the solder 16, the first heat sink 18 as the conductive member, the solder 24, the terminal 26, the solder 28, and the second heat sink 30, It is not limited to. The semiconductor device includes a semiconductor chip having an electrode, a metal base, and a conductive member having a mounting portion of the semiconductor chip and a surrounding portion surrounding the mounting portion on a surface facing the electrode, and between the electrode and the mounting portion. What is necessary is just to provide the sealing resin body which seals the solder which interposes and connects an electrode and an electroconductive member, the semiconductor chip, the at least opposing surface of a metal member, and solder. The conductive member is provided as a peripheral portion so as to surround the mounting portion, and is provided between the close contact portion where the sealing resin body is in close contact with the mounting portion and the close contact portion, and the solder is not connected, and the close contact portion It is only necessary to have a non-adhered part having lower adhesion to the sealing resin body. The present invention is not limited to a semiconductor device with a double-sided heat dissipation structure, and can also be applied to a semiconductor device with a single-sided heat dissipation structure.

Although the example of the roughening part by laser irradiation was shown as a roughening part which comprises the contact | adherence part 304, it is not limited to this. For example, the sealing resin body 14 may be brought into close contact with the contact portion 304 by roughening plating. Furthermore, as shown in the second embodiment, a resin layer that improves the adhesion with the sealing resin body 14 may be provided in the formation region of the adhesion portion 304.

According to the semiconductor device of the present disclosure, the semiconductor chip has the first electrode that is an electrode on one surface side in the thickness direction, and the second electrode on the surface side opposite to the one surface. The semiconductor device may further include a second conductive member provided so as to sandwich the semiconductor chip between the first conductive member and the second conductive member electrically connected to the second electrode. Further, the sealing resin body may integrally seal the surface of the second conductive member that faces the first conductive member. Furthermore, the first conductive member has a first contact portion and a non-contact portion which are close contact portions as a peripheral portion, and the second conductive member is in close contact with the surface facing the first conductive member. The second contact portion may be provided so as to overlap with at least a part of the non-contact portion in plan view in the thickness direction.

Further, according to this semiconductor device, even if a void is generated on the non-contact portion of the first conductive member, the void may be detected.

Heretofore, the embodiments, configurations, and aspects of the semiconductor device according to one aspect of the present disclosure have been illustrated, but the embodiments, structures, and aspects according to the present disclosure are limited to the above-described embodiments, configurations, and aspects. It is not a thing. For example, embodiments, configurations, and aspects obtained by appropriately combining technical sections disclosed in different embodiments, configurations, and aspects are also included in the scope of the embodiments, configurations, and aspects according to the present disclosure.

Claims (10)

A semiconductor chip (12) having an electrode (121);
A conductive member (18) including a metal substrate (186) and having a mounting portion (182) of the semiconductor chip and a surrounding portion (183) surrounding the mounting portion on a surface facing the electrode;
Solder (16) interposed between the electrode and the mounting portion and connecting the electrode and the conductive member;
A sealing resin body (14) for integrally sealing the semiconductor chip, at least the facing surface of the conductive member, and the solder;
With
The conductive member is provided as the peripheral portion so as to surround the mounting portion, and the conductive member includes a close contact portion (184) where the sealing resin body is in close contact, and the mounting portion and the close contact portion. A semiconductor device having a non-contact portion (185) provided therebetween, wherein the solder is not connected and adhesion to the sealing resin body is lower than that of the close-contact portion. The semiconductor device according to claim 1, wherein a width of the non-contact portion is 0.2 mm or more. 3. The semiconductor device according to claim 1, wherein a width of the non-contact portion is 0.5 mm or less. The semiconductor device according to any one of claims 1 to 3, wherein the close contact portion is a roughened portion having a continuous uneven surface. The conductive member includes a film (187) formed on the surface of the metal substrate,
The film is a metal thin film (188) formed on the surface of the metal substrate, and an oxide of the same metal as the main component metal of the metal thin film, and the surface is a concavo-convex oxide film having a concavo-convex surface ( 189), and
The concavo-convex oxide film is provided in the entire area of the close contact portion and at least the close contact portion side portion of the non close contact portion,
The semiconductor device according to claim 4, wherein a height of the convex portion in the uneven oxide film is higher in the close contact portion than in the non-contact portion. The semiconductor device according to any one of claims 1 to 3, wherein a resin layer (191) that enhances adhesion to the sealing resin body is formed only in the adhesion portion among the peripheral portions. The semiconductor chip has a first electrode (121) as the electrode on one surface side in the thickness direction, and a second electrode (122) on the surface opposite to the one surface,
A second conductive member (30) provided in the thickness direction so as to sandwich the semiconductor chip between the first conductive member, which is the conductive member, and electrically connected to the second electrode;
The sealing resin body integrally seals the surface of the second conductive member facing the first conductive member,
The first conductive member has a first contact portion and the non-contact portion which are the contact portions as the peripheral portion,
The second conductive member is a portion where the sealing resin body is in close contact with a surface facing the first conductive member, and is provided so as to overlap at least a part of the non-contact portion in plan view in the thickness direction. The semiconductor device according to any one of claims 1 to 6, further comprising a second contact portion (304) formed. The semiconductor device according to claim 7, wherein the second contact portion is provided so as to overlap an entire area of the non-contact portion. The semiconductor device according to claim 7, wherein the second contact portion is provided so as to overlap only a part in a circumferential direction of the non-contact portion surrounding the mounting portion. The semiconductor device according to any one of claims 7 to 9, wherein the second contact portion is a roughened portion having a continuous uneven surface.

Images