JP7138720B2 - Semiconductor device, power semiconductor module, power converter, and method for manufacturing power semiconductor module - Google Patents

Description

The present invention relates to technology for improving the reliability of junctions of semiconductor devices.

Most of the power semiconductor modules that make up power converters such as inverters and converters have switching devices such as IGBTs (Insulated Gate Bipolar Transistors) and MOSFETs (Metal-Oxide-Semiconductor Field-Effect Transistors) and anti-parallel switching devices. and a freewheeling diode connected to . In general, IGBTs made of silicon (Si) semiconductors are used as switching devices, and pin diodes made of Si semiconductors are used as freewheeling diodes. In recent years, semiconductor devices using silicon carbide (SiC), which is a wide bandgap semiconductor having a wider bandgap than Si semiconductors, have been developed. SiC has a dielectric breakdown strength about 10 times higher than that of Si, and the thickness of the drift layer can be reduced to about 1/10 that of Si, so that a low on-voltage can be realized and operation is possible even at high temperatures. Therefore, a semiconductor device using SiC as a semiconductor material can be made smaller and more efficient than a conventional semiconductor device using Si as a semiconductor material.

Semiconductor devices for power conversion, such as IGBTs, MOSFETs, and diodes, are of the front-to-back conduction type in which the main current flows in the thickness direction of the semiconductor device. When mounting a front-to-back conduction type semiconductor device on a ceramic substrate, the back electrode of the semiconductor device is soldered to the metal pattern of the ceramic substrate, and the front surface electrode of the semiconductor device is connected by wire bonding and soldering to the metal terminal, thereby conducting current. A path is formed. However, the cost of semiconductor devices in a semiconductor module (hereinafter simply referred to as a module) is large, and in order to reduce the size and cost of the module, the current density applied to each semiconductor device and the current density applied to the wiring are increased. tend to rise. As the current density increases, the amount of heat generated by the semiconductor device during operation increases, and the temperature of the semiconductor device and the wiring temperature rise.

Conventionally, an Al wire has been joined to a surface electrode made of an aluminum (Al) alloy of a semiconductor device. Such wiring can reduce junction life due to increased current densities and temperatures. Therefore, a module in which copper (Cu) wire, which has a higher electrical conductivity than Al wire, is applied for wiring, and a module in which a Cu plate material is soldered to the surface electrode of a semiconductor device have been developed.

The surface electrodes of the semiconductor devices of these modules are formed with Cu electrodes to improve adhesion with Cu wires, and gold (Au) is formed on nickel (Ni) plating to improve adhesion with solder. A plated electrode is formed.

For example, Patent Document 1 discloses a semiconductor device in which a metal film is formed on a surface electrode using wet plating.

JP 2008-244045 A

Patent Literature 1 discloses a semiconductor device having a configuration in which a plated electrode is formed on a surface electrode and the plated electrode and a heat sink are joined via a solder material. In Patent Document 1, in order to reduce the stress concentrated on the contact point where the protective film, the plating electrode and the solder material are in contact with each other, the distance a from the protective film to the heat sink and the side surface of the heat sink from the top surface of the plating film are wetted and spread. It defines the distance b to the top of the solder material. However, in a power semiconductor device in which the operating temperature rises and the energized current increases, the solder joints repeatedly expand and contract due to temperature swings during operation of the power semiconductor device, which causes the plated electrodes to be peeled off, resulting in the power semiconductor device. may be damaged, resulting in element failure.

SUMMARY OF THE INVENTION An object of the present invention is to solve the problems described above, and to improve the reliability of the junction of the surface electrodes of a semiconductor device.

A semiconductor device of the present disclosure includes a semiconductor substrate, a surface electrode provided on a first main surface of the semiconductor substrate, a protective film provided on the peripheral edge of the surface electrode, and a region surrounded by the protective film of the surface electrode. and a first metal film spaced apart from the protective film and arranged in contact with the surface electrode , the first metal film being a cold spray film and bonded to the plate-like wiring by solder spaced apart from the protective film. be done.

According to the present invention, since the solder used for joining the first metal film and the plate-like wiring is separated from the protective film, there is no contact point where all of the first metal film, the protective film and the solder come into contact. Therefore, the stress on the first metal film, the protective film and the solder is relieved, and the reliability of the surface electrode itself or the joint between the surface electrode and the plate-like wiring increases.

1 is a cross-sectional view of a power semiconductor module according to Embodiment 1; FIG. 2 is a partial plan view of the power semiconductor module of Embodiment 1; FIG. 2 is a top view of the switching device of Embodiment 1; FIG. FIG. 4 is a cross-sectional view taken along line CC of FIG. 3; FIG. 4 is a cross-sectional view showing a method of manufacturing the freewheeling diode of the first embodiment; FIG. 4 is a top view of a semiconductor wafer on which a protective film is formed; FIG. 2 is a top view of a semiconductor wafer with a mask overlaid on a protective film; FIG. 4 is a cross-sectional view showing a state in which powder is being discharged from the cold spray device to the upper electrode; FIG. 10 is a cross-sectional view showing a film forming process of the second cold spray film 111; BRIEF DESCRIPTION OF THE DRAWINGS It is a block diagram which shows the structure of the power conversion system to which the power converter device concerning this Embodiment is applied.

<A. Embodiment 1>
<A-1. Configuration>
FIG. 1 is a cross-sectional view of a power semiconductor module 100 (hereinafter also simply referred to as module 100) according to Embodiment 1. As shown in FIG. FIG. 2 is a partial plan view of the module 100 viewed from the top side (upper side of the page of FIG. 1). FIG. 1 is a cross-sectional view taken along line ABA in FIG. 3 is a top view of switching device 104a mounted on module 100, and FIG. 4 is a cross-sectional view taken along line CC of FIG. In FIG. 2, illustration of a cover for closing the opening, a sealing member 105, and the like is omitted. The configuration of the module 100 will be described below with reference to FIGS. 1 to 4. FIG.

As shown in FIG. 1, the module 100 includes a base plate 101, a case 102, an insulating substrate 103, a semiconductor device 104, a first cold spray film 110, a second cold spray film 111, a protective film 112, a plate wiring 106a, It has a main terminal 108a and a control terminal 108b.

Note that the cold spray film is a film formed by bonding metal particles to a surface to be bonded by colliding with the surface to be bonded and deforming metal particles ejected at a temperature within a range in which the metal particles do not melt. Since gaps are generated between the deformed and joined metal particles, the cold-sprayed film has pores inside and unevenness on the surface. The thickness and shape of the cold spray film can be arbitrarily selected.

The case 102 is open on the top side (upper side of the paper surface of FIG. 1) and the bottom side (lower side of the paper surface of FIG. 1). The base plate 101 is joined to the case 102 with a resin material or the like, and is housed in an opening on the bottom side of the case 102 . The base plate 101 has the same shape and area as the opening on the bottom side of the case 102 and constitutes the bottom surface of the case 102 . The base plate 101 is made of, for example, a Cu plate or a composite material such as AlSiC or Cu—Mo. However, the base plate 101 may not be provided if sufficient insulation performance, strength and thermal conductivity can be obtained when using the semiconductor device 104 . In this case, the lower conductor pattern 103 e of the insulating substrate 103 is exposed as the bottom surface of the case 102 . Case 102 is made of, for example, polyphenyl sulfide resin (PPS), polybutylene terephthalate resin (PBT), or polyethylene terephthalate resin (PET). By introducing the sealing material 105 such as resin from the opening on the upper surface side of the case 102, the base plate 101, the insulating substrate 103, the semiconductor device 104, the wiring 106, the main terminals 108a and the control terminals 108b are covered with the sealing material 105. covered. For the sealing material 105, for example, an epoxy material mixed with a filler or silicon gel is used. The sealing material 105 only needs to have sufficient insulation when the module 100 is used.

The lower conductor pattern 103e of the insulating substrate 103 is joined to the upper surface of the base plate 101 with solder 107b. The insulating substrate 103 includes an insulating material 103d, upper conductor patterns 103a, 103b, and 103c formed on the upper surface of the insulating material 103d, and a lower conductor pattern 103e formed on the lower surface of the insulating material 103d. The insulating material 103d is a ceramic material such as aluminum oxide (Al ₂ O ₃ ), aluminum nitride (AlN) or silicon nitride (Si ₃ N ₄ ), or a binder material such as an epoxy material or a liquid crystal polymer containing silica, alumina or nitride. It is composed of an organic insulating material mixed with a filler such as boron (BN). The upper conductor patterns 103a, 103b, 103c and the lower conductor pattern 103e are, for example, a single Cu material, a Cu material plated with Ni or silver (Ag), or an Al material plated with Ni or Ag. consists of

A semiconductor device 104 is bonded to the upper surface of the insulating substrate 103 with solder 107a. The solder 107a is a solder material whose base material is tin (Sn) or lead (Pb), for example. Alternatively, an Ag sintered material may be used instead of the solder 107a. The semiconductor device 104 includes a switching device 104a and a freewheeling diode 104b. The switching device 104a is, for example, SiC-MOSFET or Si-IGBT, and the freewheeling diode 104b is, for example, SiC-SBD (Shottky Barrier Diode) or Si-FWD (Free Wheeling Diode). The upper conductor pattern 103a is electrically connected to the drain electrode of the switching device 104a. In the following description, the switching device 104a and the freewheeling diode 104b are simply referred to as the semiconductor device 104 when there is no need to distinguish between them.

The case 102 is provided with a main terminal 108a and a control terminal 108b. The switching device 104a is connected to the main terminal 108a by the plate-like wiring 106a, and is connected to the control terminal 108b by the source signal wiring 106b and the gate signal wiring 106c. The freewheeling diode 104b is connected by a main terminal 108a and a plate-like wiring 106a.

Although not shown in FIG. 1, as shown in FIGS. 2 to 4, upper surface electrodes 109 are provided on the upper surfaces of the switching device 104a and the freewheeling diode 104b. Furthermore, a gate pad 109a is provided on the upper surface of the switching device 104a. As shown in FIGS. 3 and 4, a protective film 112 is formed on the periphery of the upper surface of the switching device 104a. The protective film 112 is also formed on the upper surface of the switching device 104a so as to surround the gate pad 109a. That is, the upper electrode 109 and the gate pad 109a are insulated by the protective film 112. FIG. The protective film 112 is provided for physical protection of the surface of the switching device 104a and for increasing the insulation distance. For the protective film 112, SiO ₂ or SiN, which is an inorganic material, is used in addition to polyimide, which is an organic material.

A third cold spray film 113 is formed on the gate pad 109a. The third cold spray film 113 is connected to the upper conductor pattern 103c by wedge bonding through the gate signal wiring 106c. The upper conductor pattern 103c is electrically connected to the gate electrode, and is also connected to the control terminal 108b and the gate signal wiring 106c. Here, the gate signal wiring 106c is a Cu wire, and the third cold spray film 113 is used as a junction buffer layer for the Cu wire. Therefore, when other materials such as Au wire are used for the gate signal wiring 106c, the third cold spray film 113 need not be formed on the gate pad 109a. However, when a Cu wire is used for the gate signal wiring 106c, the reliability of the junction is significantly higher than that of an Al wire. Since Cu has a higher strength than Al, deformation of the wire is less likely to occur during assembly. In this case, the wire diameter of the gate signal wiring 106c is, for example, about 50 μm or more and 200 μm or less.

As shown in FIG. 2, the upper electrode 109 of the switching device 104a is connected by wedge bonding with the upper conductor pattern 103b and the source signal wiring 106b. Upper conductor pattern 103b is electrically connected to the source electrode. A conductive material may be used for the source signal wiring 106b, such as a Cu wire or an Al wire.

Although not shown in FIGS. 1 and 2, the module 100 is provided with a diode for temperature detection. A Cu wire or an Al wire is also used for the wiring connected to the cathode electrode and the anode electrode of this diode, like the gate signal wiring 106c.

As described above, the switching device 104a and the freewheeling diode 104b are connected to the main terminal 108a by the plate-like wiring 106a. Next, a junction between the switching device 104a and the freewheeling diode 104b and the plate-like wiring 106a will be described. The junction between the switching device 104a and the plate-like wiring 106a will be described below with reference to FIG.

A plate material made of Cu, for example, is used for the plate-like wiring 106a. However, the plate-like wiring 106a may be made of a plate material having a smaller coefficient of thermal expansion than Cu, as long as it has electrical conductivity that allows current to flow. For example, the plate-like wiring 106a may be a plate material having a Cu--Invar--Cu layer structure in which front and back surfaces of Invar, which is an Fe--Ni alloy, are plated with Cu. The plate-shaped wiring 106a is joined to the main terminal 108a by, for example, welding, soldering, or ultrasonic joining. Specifically, the plate thickness of the plate-like wiring 106a is determined by the current to be supplied, and is suitably about 0.3 mm or more and 1.5 mm or less.

A first cold spray film 110 is formed by a cold spray method on the outermost surface of the upper electrode 109 of the switching device 104a and the freewheeling diode 104b. The first cold spray film 110 is also called a first metal film. Plate-shaped wirings 106a are provided at positions facing each other on these first cold spray films 110, and both are joined by solder 107c. The first cold spray film 110 is, for example, a cold spray film of Cu powder, but may be a cold spray film of other metal powder as long as it is a metal film that allows the solder 107c to wet and spread. It does not have to be a cold spray film.

It is desirable that the first cold spray film 110 is formed while avoiding the end portion of the semiconductor device 104 so as not to come into contact with the protective film 112 . Accordingly, even if the solder 107c wets and spreads on the first cold spray film 110, no contact point is formed where the solder 107c, the protective film 112 and the first cold spray film 110 are in contact with each other. However, if the solder 107c can be prevented from coming into contact with the protective film 112 by the second cold spray film 111, as will be described later, the protective film 112 may be in contact with the first cold spray film 110, or the second cold spray film 111 may be in contact with the second cold spray film 110. It may be in contact with the spray film 111 . A method for forming the first cold spray film 110 and the second cold spray film 111 will be described later.

As shown in FIG. 1, a through hole is formed at the position of the plate-like wiring 106a facing the first cold spray film 110. As shown in FIG. This through-hole functions as a space in which the solder 107c provided in the space between the first cold spray film 110 and the plate-like wiring 106a can wet and spread when it melts and expands in volume. As a result, the solder that does not fit between the first cold spray film 110 and the plate-like wiring 106a is prevented from becoming solder balls and being separated. As shown in FIG. 1, it is desirable that the peripheral portion of the through-hole of the plate-like wiring 106a protrude downward with respect to the other portions. Such a shape can be obtained by pressing the plate-shaped wiring 106a downward and then forming a through-hole having a diameter smaller than the pressed region by pressing.

If the solder 107c is thick, the probability of solder ball formation during melting increases. On the other hand, if the solder 107c is thin, the stress cannot be relieved when the plate-shaped wiring 106a is deformed due to thermal expansion, and cracks are generated from the first cold spray film 110 to the upper electrode 109. less reliable. Therefore, the thickness of the solder 107c is preferably about 400 μm. When punching out the through holes of the plate-like wiring 106 a by pressing, it is desirable to punch out in the direction facing the first cold spray film 110 .

A second cold spray film 111 is formed on the periphery of the first cold spray film 110 . The second cold spray film 111 is also called a second metal film. The second cold spray film 111 is in contact with the plate-like wiring 106a and serves as a wall to prevent the melted solder 107c from spreading onto the upper electrode 109. FIG. As a result, even if the amount of solder 107c supplied is too large to fit between the first cold spray film 110 and the plate-like wiring 106a, the solder 107c passes through the through-holes of the plate-like wiring 106a to the plate-like wiring 106a. Wetting and spreading to the upper part of the solder ball suppresses the occurrence of solder balls. The second cold spray film 111 is made of, for example, a Cu material, but any metal material that allows the molten solder 107c to wet and spread may be used, and the powder supplied to the first cold spray film 110 may be different. That is, when the melted solder 107c does not remain inside the second cold spray film 111, it spreads on the outer side surface of the second cold spray film 111, thereby suppressing the formation of solder balls. Even when the first cold spray film is wetted with the solder, it is possible to suppress the formation of solder balls. The thickness of the first cold spray film 110 is preferably 50 μm or more and less than 200 μm, and the thickness of the second cold spray film 111 is preferably about 200 μm or more and less than 1000 μm. The second cold spray film 111 is desirably made of a material that has better wettability with respect to the solder 107c than the first cold spray film 110 does. Note that even if the first cold spray film 110 is made of a material having better wettability than the second cold spray film 111, the effect of suppressing solder balls can be obtained in the same manner.

As a method of supplying the solder 107c when assembling the module 100, a plate-like solder material may be sandwiched between the first cold spray film 110 and the plate-like wiring 106a. Alternatively, a cylindrical solder material having a diameter smaller than that of the through-hole of the plate-like wiring 106a may be supplied from above the through-hole.

Thus, the first cold spray film 110 is formed on the upper electrode 109 of the semiconductor device 104 and the second cold spray film 111 is formed on the peripheral edge of the first cold spray film 110 . The contact of the second cold spray film 111 with the plate-like wiring 106 a suppresses the melted solder 107 c from overflowing onto the upper electrode 109 . In addition, since the melted solder 107c remains in the through-hole of the plate-like wiring 106a, the solder 107c is prevented from spilling from the upper surface of the plate-like wiring 106a to form a solder ball. Moreover, since the solder 107c formed in the through-hole of the plate-like wiring 106a serves as a rivet, deformation of the plate-like wiring 106a during thermal expansion is suppressed. Also, if there is a gap between the second cold spray film 111 and the plate-like wiring 106a, the solder 107c may spill from this gap. is wetted and spread, suppressing the occurrence of solder balls.

When the second cold spray film 111 is in contact with the plate-like wiring 106 a , the solder 107 c is covered with the second cold spray film 111 between the plate-like wiring 106 a and the semiconductor device 104 and does not contact the sealing material 105 . By reducing the contact area between the solder 107c and the encapsulating material 105 in this way, even if the encapsulating material 105 is made of an epoxy resin having poor adhesion to the solder 107c, the solder 107c of the encapsulating material 105 does not reach the solder 107c. Peeling is suppressed. In addition, when the first cold spray film 110 and the second cold spray film 111 are made of a Cu material, the anchor effect of the unevenness of the surface provides high adhesion to the epoxy resin. further suppressed.

As shown in FIG. 4, the switching device 104a comprises a semiconductor substrate 201, a top electrode 109, a gate pad 109a, a barrier metal 204, a Ni electrode 205 and an Au electrode 206. The semiconductor substrate 201 is a SiC substrate or a Si substrate, and has a first main surface 201a and a second main surface 201b facing the first main surface 201a. Diffusion layers (not shown) are provided on the surface layers on the first main surface 201a side and the second main surface 201b side of the semiconductor substrate 201, and these diffusion layers function as active layers that control the operation of the MOSFET or IGBT. .

Upper electrode 109 and gate pad 109 a are formed on first main surface 201 a of semiconductor substrate 201 . That is, the upper surface electrode 109 and the gate pad 109a are surface electrodes provided on the first main surface 201a of the semiconductor substrate 201 . The upper electrode 109 is a source electrode when the semiconductor device 104 is a MOSFET, and is an emitter electrode when the semiconductor device 104 is an IGBT. The upper surface electrode 109 and the gate pad 109a are electrodes for electrical conduction with the diffusion layer of the surface layer on the first main surface 201a side of the semiconductor substrate 201, and generally pure Al, AlSi alloy, AlCu alloy and AlSiCu alloy are used. . The mixing ratio of Si or Cu in the alloy is 5 wt % or less by weight in the alloy. The upper surface electrode 109 and the gate pad 109a form a first main electrode 210 together with a surface diffusion layer on the side of the first main surface 201a of the semiconductor substrate 201 .

A barrier metal 204 is provided on the second main surface 201 b of the semiconductor substrate 201 . The barrier metal 204 is a film that prevents entry of elements that inhibit device operation from the outside, and is generally titanium (Ti), molybdenum (Mo), tungsten (W), vanadium (V), chromium (Cr) or Al. metal films such as , or oxides or nitrides thereof are used. A Ni electrode 205 is directly bonded onto the barrier metal 204 . The Ni electrode 205 reacts with the solder during soldering to form alloy layers and intermetallic compounds. An Au electrode 206 is provided on the Ni electrode 205 . The Au electrode 206 is provided to eliminate the influence of contact resistance when measuring the characteristics of the semiconductor device, and to prevent oxidation of Ni and improve adhesion of solder. The barrier metal 204 , the Ni electrode 205 and the Au electrode 206 constitute a second main electrode 220 together with a surface diffusion layer (not shown) on the second main surface 201 b side of the semiconductor substrate 201 . The barrier metal 204, the Ni electrode 205 and the Au electrode 206 are drain electrodes when the semiconductor device 104 is a MOSFET, and collector electrodes when the semiconductor device 104 is an IGBT.

<A-2. Manufacturing method>
The first cold spray film 110 is formed on the upper surface electrode 109 and the gate pad 109a, which are surfaces to be bonded, by ejecting metal powder from a cold spray device. When the metal powder discharged from the cold spray device collides with the upper electrode 109 and the gate pad 109a, which are the surfaces to be bonded, the thermal energy and kinetic energy of the metal powder change into frictional heat, and the metal powder and the surface to be bonded collide. Deformation occurs on the respective outermost surfaces and they are entangled with each other. In addition, the metal oxide film on each outermost surface is removed, and a cold spray film is formed by the reaction occurring at the interface between the metal powder and the surface to be joined. Metal powder is continuously ejected from the cold spray device and collides with the previously formed cold spray film, thereby increasing the film thickness and finally forming the first cold spray film 110 .

The metal powder supplied to the cold spray device is Cu powder, and the particle size is determined by passing through a test sieve with an opening of 100 μm (149 mesh) conforming to JIS (Japanese Industrial Standards). controlled. As a method for producing Cu powder, an electrolysis method, a high-pressure swirling water atomization method, a water atomization method, or the like is used. The particle shape of the Cu powder produced by the electrolytic method is close to a dendrite shape, and the particle shape of the Cu powder produced by the high-pressure swirling water atomization method is close to a spherical shape. Regardless of the shape of the Cu powder, when it collides with the surface to be joined, it becomes a thin flat shape that does not retain the original shape.

Particles passing through a sieve with an opening of 100 μm have a particle size distribution of 100 μm or less. However, when measuring the particle size by analyzing the diffraction scattering pattern by the laser diffraction method, in the particle size distribution chart with the horizontal axis as the particle size and the vertical axis as the frequency (number), the particle with the maximum frequency (number) It is desirable that the diameter is about 20 μm. That is, it is desirable that particles with a particle size of about 20 μm form a distribution peak, and particles with smaller and larger particle sizes are present around the particles forming the peak. If anything, it is desirable to have as many particles with a particle size of 20 μm or less as possible from the viewpoint of obtaining a cold spray film with dense particle spacing, and a metal powder with such a distribution can be obtained. As such, it is desirable to control the particle production conditions. When the particle size peak is 20 μm, not only particles with a particle size of 20 μm but also particles with a particle size having a difference of ±10% from 20 μm are considered to form a peak.

With such a particle size distribution, the powder particle spacing of the first cold spray film 110 is made dense on the upper electrode 109 side and the gate pad 109a side, and becomes coarser away from the upper electrode 109 and the gate pad 109a, It can be densified again as it approaches the topmost surface of the first cold spray film 110 . More on this later.

Note that the opening of the sieve may be less than 100 μm. For example, when a sieve with an opening of 53 μm (270 mesh) is used, particles with a particle size of 53 μm or less can be obtained. However, since the yield is low, it is desirable that the opening of the sieve is about 100 μm in order to form the cold spray film at low cost.

Particles with a large particle size cannot obtain sufficient kinetic energy to form a cold spray film when ejected from a nozzle of a cold spray device, which will be described later. Particles do not bond to each other and are repelled out of the film by the flow of discharged gas, and do not contribute to film formation. Therefore, in order to uniformly form the first cold-sprayed film 110, the particle size of the metal powder is preferably as small as 100 μm or less, more preferably 60 μm or less. Although the lower limit of the particle size of the metal powder is, for example, 1.0 μm, it is not necessary to perform screening with a sieve to control the lower limit of the particle size. In addition, for the reasons explained below, it is not necessary to make the particle size of the metal powder uniform, and it is not necessary to perform screening by a sieve to make the particle size uniform.

That is, the metal powder passed through a sieve having a predetermined mesh size has a particle size distribution equal to or less than the mesh size. When film formation is performed by cold spray using such metal powder, finer particles with lighter weight preferentially collide with the upper electrode 109 and the gate pad 109a to form a cold spray film. Particles of heavy weight and large size impinge on the previously formed cold spray film. Therefore, switching device 104a protected by upper electrode 109 and gate pad 109a is not destroyed. Furthermore, since fine particles collide with the upper electrode 109 and the gate pad 109a in the initial stage of film formation, the lower surface side of the first cold spray film 110, that is, the side closer to the upper electrode 109 and the gate pad 109a, has a smaller particle size. It results in a dense cold spray film composed of small particles. Since the metal powder having a particle size distribution equal to or smaller than the above-described mesh size collides successively with the dense cold spray film, a coarser cold spray film is formed on the dense cold spray film. In addition, since particles with a small particle size collide with each other between particles with a large particle size, a film with a relatively small interval between powder particles is formed, so the upper layer of the first cold spray film 110 is a relatively dense film. becomes. As a result, the first cold spray film 110 has a film structure in which the distance between the powder particles becomes smaller toward the upper electrode 109 and the gate pad 109a, becomes larger as it separates from the upper electrode 109 and the gate pad 109a, and becomes smaller as it approaches the outermost surface. have

The thickness of the upper electrode 109 and the gate pad 109a is, for example, 5 μm. However, the upper electrode 109 and the gate pad 109a may be thinner or thicker than 5 μm as long as the damage to the semiconductor device due to the deposition of the first cold spray film 110 can be prevented.

On the upper electrode 109 and the gate pad 109a, a Ni plating film may be formed by wet plating, or a Ni sputter film may be formed by vacuum deposition.

In this embodiment, Cu powder is used as the metal powder that is the material of the first cold spray film 110 . However, the metal powder may be, for example, Ni powder as long as it is a metal that forms a metal bond with the surface to be bonded and serves as a material to be bonded when the plate-like wiring 106a is soldered.

The thickness of the first cold spray film 110 can be controlled by spray irradiation time, irradiation speed (gas pressure) and irradiation temperature. In this embodiment, the first cold spray film 110 does not need to be dissolved in the molten solder, and from that point of view, a thickness of 5 μm or more is sufficient. Also, if the first cold spray film 110 is too thick, the semiconductor wafer on which the switching device 104a is formed will warp and crack. Therefore, the thickness of the first cold spray film 110 is desirably 100 μm or less.

The crystal size of the Cu film formed by wet plating is about 5 μm, while the first cold spray film 110 is mainly formed of particles of about 20 μm. Therefore, the outermost surface of the first cold-sprayed film 110 is not smooth as compared with the Cu film formed by wet plating, but is uneven and uneven. The unevenness preferably has an arithmetic mean roughness Ra of 1 μm or more and 10 μm or less derived from line roughness measured in a range of 100 μm×100 μm with a laser microscope, for example. The size of the unevenness here is measured by a non-contact white light interferometer.

In this way, the presence of non-uniform unevenness on the outermost surface of the first cold spray film 110 increases the area over which the solder 107c wets and spreads, thereby improving the bonding strength between the solder 107c and the first cold spray film 110. do. Also, the adhesion between the first cold spray film 110 and the sealing material 105 is improved due to the anchor effect.

The particle size in the cold spray film described above is, for example, a value measured by observing the cross section of the first cold spray film 110 and using an electron back scatter diffraction pattern (EBSD).

The metal powder discharged from the nozzle of the cold spray device is deformed when it collides with the surfaces to be joined. Therefore, when forming the first cold spray film 110 with a film thickness of 30 μm, for example, the particle size of the metal powder does not necessarily have to be 30 μm or less.

Usually, when a metal film such as Cu is formed by a sputtering vapor deposition method, the film thickness is 2 μm or more and 3 μm or less, and the film thickness distribution is flat on the order of nanometers. When the metal film is formed by plating, the film thickness is several μm or more and 30 μm or less, and the film thickness distribution is flat on the order of submicrons. On the other hand, as will be described later, in this embodiment, a metal film is formed by a cold spray method using a mask. The cold spray film does not rise vertically along the wall surface of the mask opening, but has a shape that is inclined toward the central portion of the film. The thickness of the first cold spray film 110 can be measured by ultrasonic testing. That is, the time required for the ultrasonic wave emitted from the ultrasonic probe to pass through the first cold spray film 110 to the semiconductor device after the first cold spray film 110 is formed, and the time required for the first cold spray film 110 and the upper surface It is measured by the difference from the time until the ultrasonic wave is reflected at the interface between the electrode 109 and the gate pad 109a.

Next, a method of manufacturing the freewheeling diode 104b will be described with reference to FIGS.

As shown in FIG. 5, a top electrode 109 is formed on the first main surface 201a of the semiconductor substrate 201 provided with a diffusion layer (not shown). The diffusion layer on the side of the first main surface 201 a of the semiconductor substrate 201 and the upper surface electrode 109 form the first main electrode 210 . The top electrode 109 is formed by a vacuum deposition method such as physical vapor deposition (PVD) or chemical vapor deposition (CVD). After that, the semiconductor substrate 201 is ground from the second main surface 201b side to have a predetermined thickness.

Next, a barrier metal 204, a Ni electrode 205, and an Au electrode 206 are formed in this order on the ground second main surface 201b by, for example, a vacuum film forming method, and these are used as the second main electrode 220. FIG.

Next, a protective film 112 having a thickness of 2 μm or more and 10 μm or less is formed by applying photosensitive polyimide (PI) on the upper electrode 109 and selectively exposing it. When the protective film 112 is composed of an organic substance, it is formed after forming the Ni electrode 205 and the Au electrode 206 . After that, a cold spray mask 404 is overlaid on the protective film 112 .

A portion of the mask 404 overlapping the protective film 112 has a thickness of about 10 μm so as not to interfere with the protective film 112 . The opening of the mask 404 is arranged so as to overlap the upper electrode 109 .

FIG. 6 is a top view of the semiconductor wafer 301 on which the protective film 112 is formed. A semiconductor wafer 301 corresponds to the semiconductor substrate 201 in FIG. As shown in FIG. 6 , the protective film 112 is provided along the dicing lines 304 in the rectangular semiconductor device forming region 302 surrounded by the dicing lines 304 and is not formed on the dicing lines 304 .

FIG. 7 is a top view of semiconductor wafer 301 with mask 404 overlaid on protective film 112 . A mask 404 covers the protective film 112 and the dicing lines 304 . The opening OP of the mask 404 is the portion where the first cold spray film 110 of the semiconductor device formation region 302 is formed. A stainless steel (SUS) plate, for example, is used for the mask 404 . The thickness of the mask 404 is set so that the powder discharged by the cold spray device does not penetrate the mask 404, for example, 0.5 mm or more and less than 5 mm, preferably about 1 mm. The mask 404 may be a mask 404 in which the portion of the SUS plate facing the protective film 112 is coated with the same organic substance as the protective film 112 . This reduces the possibility that the protective film 112 is damaged due to direct contact between the protective film 112 and the SUS plate.

FIG. 8 shows a state in which powder 423 is being discharged from the cold spray device 420 to the upper electrode 109 . A cold spray device 420 is arranged above the semiconductor wafer 301 on which the mask 404 is superimposed. A first cold spray film 110 is deposited on the electrode 109 . The cold spray device 420 comprises a spray gun 421 , a nozzle 422 and a pipe 424 . Nozzle 422 is provided at the tip of spray gun 421 to eject powder 423 . The pipe 424 is provided on the side of the spray gun 421 opposite to the nozzle 422 . The powder 423 is metal powder having a particle size of 100 μm or less, and in the present embodiment, Cu powder is used as a material that can be soldered to a Cu plate that is a plate-shaped wiring material. The powder 423 may be metal powder such as Ni, Ag, Pd, and Au.

The cold spray device 420 is connected via piping 424 to a powder feeder (not shown) and a high pressure gas generator (not shown). Powder 423 supplied from the powder supply device to pipe 424 is heated to a high temperature when passing through spray gun 421 fitted with a heater (not shown). A high-pressure gas is supplied to the pipe 424 from the high-pressure gas generator to energize the powder 423 heated to a high temperature. As a result, powder 423 is discharged from nozzle 422 to upper electrode 109 .

The high-pressure gas may be any inert gas, such as nitrogen gas, argon gas or helium gas. When the pressure of the high-pressure gas is discharged from the tip of the nozzle 422, for example, the injection pressure is 0.1 MPa or more and 10 MPa or less, the flow rate is 1 ml/min or more and 1000 ml/min or less, and the powder supply amount is 1 g/min or more and 100 g. /min or less. By spraying the powder 423 under such conditions, it is possible to form the first cold spray film 110 with an arbitrary film thickness according to the discharge time. Further, by setting the pressure as described above, as described above, finer particles are preferentially caused to collide with the upper electrode 109 to form the first cold spray film 110. Large particles can impinge on the previously formed first cold spray film 110 .

Note that the heater temperature can be adjusted in the range of 100 to 700° C., and is arbitrarily selected according to the material to be discharged.

If the distance between the tip of the nozzle 422 and the mask 404 is 1 mm or more and 300 mm or less, the first cold spray film 110 can be formed. In this embodiment, the distance is set to about 5 mm. This distance is a distance for excluding particles having a large particle size, as explained above. Particles that are too large to obtain sufficient kinetic energy to form the first cold spray film 110 when ejected from the nozzle 422 are ejected from between the tip of the nozzle 422 and the mask 404 in the ejected gas. It will be repelled out of the membrane by the flow.

After forming the first cold spray film 110, a second cold spray film 111 is formed as shown in FIG. First, powder that did not contribute to film formation is removed from the semiconductor wafer 301 with a blower. Here, the application width of the cold spray by the nozzle 422 is set to 500 μm or more and 1000 μm or less. Next, the center of the nozzle 422 is set inside about 250 μm from the opening edge of the mask 404 . Then, the powder 423 is discharged from the nozzle 422 and cold sprayed around the opening of the mask 404 . Thereby, a second cold spray film 111 is formed on the edge of the first cold spray film 110 . The thickness of the second cold spray film 111 peaks at the center position of the nozzle 422, and the thickness of the second cold spray film 111 becomes smaller toward the center of the free wheel diode 104b. When the first cold spray film 110 and the second cold spray film 111 are viewed as an integrated cold spray film, the cold spray film has a concave cross section. Here, the thickness of the second cold spray film 111 can be arbitrarily set within a range smaller than the distance between the upper electrode 109 and the plate-like wiring 106a.

After that, the mask 404 is removed, and the powder that has not contributed to the film formation is removed from the semiconductor wafer 301 with a blower. After that, the first cold spray film 110 and the second cold spray film 111 are formed by ultrasonic cleaning in a reducing liquid or in an acidic liquid that slightly dissolves Cu so as not to oxidize. Remove any powder that did not contribute to the membrane. Alternatively, the oxide film formed on the outermost surface of the first cold spray film 110 may be removed, and then an antirust film such as benzotriazole may be formed to prevent oxidation.

Further, the semiconductor wafer 301 is dried in an inert gas atmosphere such as nitrogen gas to relieve the residual stress generated in the first cold spray film 110 and the second cold spray film 111 when the powder 423 collides. , the bonding interfaces between particles are recrystallized to form a denser metal film.

The semiconductor wafer 301 on which the first cold spray film 110 and the second cold spray film 111 are formed is diced by a dicing machine along dicing lines 304, and singulated into free wheel diodes 104b as shown in FIG. be done.

Drying and recrystallization with nitrogen gas may be performed after the dicing process instead of before the dicing process. For example, during the process of assembling the module 100, more specifically, during chip die bonding, reflow heat is used to relax the residual stress and recrystallize the first cold spray film 110 and the second cold spray film 111. Efficient.

<A-3. Variation>
In the above description, it has been described that the first cold spray film 110 and the second cold spray film 111 are provided on the upper surface electrode 109 and the gate pad 109a. However, instead of at least one of the first cold spray film 110 and the second cold spray film 111, a Ni plating film may be formed.

<A-4. Effect>
The semiconductor device 104 of the first embodiment includes a semiconductor substrate 201, an upper surface electrode 109 or a gate pad 109a which is a surface electrode provided on a first main surface 201a of the semiconductor substrate 201, and a peripheral edge of the surface electrode. and a first cold spray film 110 which is a first metal film provided on a region surrounded by the protective film 112 of the surface electrode. It is joined to the plate-like wiring 106a by the solder 107c spaced apart from the . That is, there is no contact point where all of the first cold spray film 110, protective film 112 and solder 107c are in contact. This avoids concentration of stress on the first cold spray film 110, the protective film 112 and the solder 107c. Therefore, the reliability of the surface electrode itself, or the first cold spray film 110 and the solder 107c, which is the joint between the surface electrode and the plate-like wiring 106a, is enhanced. Specifically, crack propagation in the solder 107c, the first cold spray film 110, and the surface electrode is suppressed. Further, even if a crack occurs in the solder 107c and the crack propagates to the first cold spray film 110, the crack propagation distance from the solder 107c to the surface electrode is long by the thickness of the first cold spray film 110. Long life is achieved. Moreover, it is possible to prevent the semiconductor substrate 201 from being peeled off from the contact point of the surface electrode with the protective film 112 and destroyed. When the semiconductor device is a MOSFET or IGBT, if a crack occurs in the surface electrode, the crack grows vertically and reaches the vitreous gate oxide film, causing a short-circuit failure between the gate and the source or between the gate and the emitter. put away. However, according to the semiconductor device 104, cracks in the surface electrode 104 can be suppressed, so that the above short circuit failure can be suppressed.

In addition, since the first cold spray film 110 is spaced apart from the protective film 112, that is, does not come into contact with the protective film 112, even if the solder 107c spreads beyond the first cold spray film 110, the first cold spray film 110 is , the protective film 112 and the solder 107c do not all come into contact with each other, so that the peeling of the solder 107c is suppressed and stress concentration on the surface electrode is avoided. Therefore, the reliability of the surface electrode itself, or the first cold spray film 110 and the solder 107c, which is the joint between the surface electrode and the plate-like wiring 106a, is enhanced.

The semiconductor device 104 of the first embodiment is a second metal film provided on the peripheral edge of the first cold spray film 110 between the first cold spray film 110, which is the first metal film, and the plate-like wiring 106a. A second cold spray film 111 is provided. Therefore, the second cold spray film 111 can prevent the solder 107c from spilling over the surface electrode and becoming a solder ball.

When the first metal film is the first cold spray film 110 formed by the cold spray method, the solder 107c is more uneven than the metal film formed by wet plating. The area that wets and spreads over the top is large, and the bonding strength with the solder 107c is improved.

When the second metal film is the second cold spray film 111 formed by the cold spray method, the solder 107c is more uneven than the metal film formed by wet plating. The area that wets and spreads over the top is large, and the bonding strength with the solder 107c is improved. In addition, since the outer side surface of the second cold spray film 111 has unevenness, the adhesion between the second cold spray film 111 and the sealing material 105 is improved by an anchor effect.

The power semiconductor module 100 of the first embodiment includes a semiconductor device 104 and a plate-shaped wiring 106a provided at a position facing a first metal film of the semiconductor device 104 and joined to the first metal film with solder 107c. Prepare. Since the power semiconductor module 100 does not have a contact point where all of the first metal film, the protective film 112 and the solder 107c contact, the surface electrode itself or the junction between the surface electrode and the plate-like wiring 106a Certain first cold spray films 110 and solder 107c are more reliable.

In the power semiconductor module 100 of the first embodiment, when a through hole is formed in a position facing the first metal film of the plate-like wiring 106a, it is provided in the space between the first metal film and the plate-like wiring 106a. When the solder 107c melts and expands in volume, it can wet and spread into the through-hole, so solder balls can be suppressed.

In the power semiconductor module 100 of the first embodiment, if there is a gap between the second metal film and the plate-like wiring 106a, the solder 107c may spill from this gap. Since the solder 107c spreads over the surface of the spray film 111, the occurrence of solder balls is suppressed.

In the method of manufacturing the power semiconductor module 100 of the first embodiment, the upper surface electrode 109 or the gate pad 109a, which is a surface electrode, is formed on the first main surface 201a of the semiconductor substrate 201, and the protective film 112 is formed on the periphery of the surface electrode. is formed, the first cold spray film 110, which is the first metal film, is formed on the area surrounded by the protective film 112 of the surface electrode by the cold spray method, and the first metal film is separated from the protective film 112 by the solder It is joined to the plate-like wiring 106a by 107c. This avoids concentration of stress on the first cold spray film 110, the protective film 112 and the solder 107c. Therefore, the reliability of the surface electrode itself, or the first cold spray film 110 and the solder 107c, which is the joint between the surface electrode and the plate-like wiring 106a, is enhanced.

<B. Embodiment 2>
The present embodiment applies the module 100 according to the first embodiment described above to a power converter. Although the module 100 according to the first embodiment is not limited to a specific power converter, a case where the module 100 according to the first embodiment is applied to a three-phase inverter will be described below.

FIG. 10 is a block diagram showing the configuration of a power conversion system to which the power conversion device according to this embodiment is applied.

The power conversion system shown in FIG. 10 is composed of a power supply 500, a power conversion device 510 and a load 520. The power supply 500 is a DC power supply and supplies DC power to the power converter 510 . The power supply 500 can be composed of various things, for example, it can be composed of a DC system, a solar battery, a storage battery, or it can be composed of a rectifier circuit or an AC/DC converter connected to an AC system. good too. Also, power supply 500 may be configured by a DC/DC converter that converts DC power output from a DC system into predetermined power.

Power converter 510 is a three-phase inverter connected between power supply 500 and load 520 , converts DC power supplied from power supply 500 into AC power, and supplies AC power to load 520 . As shown in FIG. 10, the power conversion device 510 includes a main conversion circuit 511 that converts DC power into AC power and outputs it, and a control circuit 513 that outputs a control signal for controlling the main conversion circuit 511 to the main conversion circuit 511. and

Load 520 is a three-phase electric motor driven by AC power supplied from power conversion device 510 . Note that the load 520 is not limited to a specific application, but is an electric motor mounted on various electrical equipment, such as a hybrid vehicle, an electric vehicle, a railway vehicle, an elevator, or an electric motor for air conditioning equipment.

Details of the power conversion device 510 will be described below. The main conversion circuit 511 includes a switching element and a freewheeling diode (not shown). By switching the switching element, the DC power supplied from the power supply 500 is converted into AC power and supplied to the load 520 . Although there are various specific circuit configurations of the main conversion circuit 511, the main conversion circuit 511 according to the present embodiment is a two-level three-phase full bridge circuit, and has six switching elements and It can consist of six freewheeling diodes in anti-parallel. A semiconductor module 512 corresponding to the module 100 of the first embodiment described above is applied to at least one of each switching element and each freewheeling diode of the main converter circuit 511 . Six switching elements are connected in series every two switching elements to form upper and lower arms, and each upper and lower arm forms each phase (U phase, V phase, W phase) of the full bridge circuit. Output terminals of the upper and lower arms, that is, three output terminals of the main conversion circuit 511 are connected to a load 520 .

Further, the main conversion circuit 511 includes a drive circuit (not shown) for driving each switching element. It may be a configuration provided. The drive circuit generates a drive signal for driving the switching element of the main conversion circuit 511 and supplies it to the control electrode of the switching element of the main conversion circuit 511 . Specifically, in accordance with a control signal from the control circuit 513, which will be described later, a drive signal for turning on the switching element and a drive signal for turning off the switching element are output to the control electrode of each switching element. When maintaining the switching element in the ON state, the driving signal is a voltage signal (ON signal) equal to or higher than the threshold voltage of the switching element, and when maintaining the switching element in the OFF state, the driving signal is a voltage equal to or less than the threshold voltage of the switching element. signal (off signal).

The control circuit 513 controls the switching elements of the main conversion circuit 511 so that desired power is supplied to the load 520 . Specifically, based on the power to be supplied to the load 520, the time (on time) during which each switching element of the main conversion circuit 511 should be in the ON state is calculated. For example, the main conversion circuit 511 can be controlled by PWM control that modulates the ON time of the switching element according to the voltage to be output. Then, a control command (control signal) to the drive circuit provided in the main conversion circuit 511 so that an ON signal is output to the switching element that should be in the ON state at each time point, and an OFF signal is output to the switching element that should be in the OFF state. to output The drive circuit outputs an ON signal or an OFF signal as a drive signal to the control electrode of each switching element according to this control signal.

In the power converter according to the present embodiment, since the module 100 according to the first embodiment is applied as the switching element and freewheel diode of the main converter circuit 511, high reliability can be obtained.

In this embodiment, an example in which the module 100 is applied to a two-level three-phase inverter has been described, but the module 100 is not limited to this and can be applied to various power converters. In this embodiment, a two-level power converter is used, but a three-level or multi-level power converter may be used. You can apply it. Also, when power is supplied to a DC load or the like, the module 100 can be applied to a DC/DC converter or an AC/DC converter.

In addition, the power conversion device to which the module 100 is applied is not limited to the case where the above-described load is an electric motor. It can also be used as a device, and can also be used as a power conditioner for a photovoltaic power generation system, an electric storage system, or the like.

In addition, within the scope of the invention, each embodiment can be freely combined, and each embodiment can be appropriately modified or omitted. Although the present invention has been described in detail, the above description is, in all its aspects, exemplary, and the invention is not limited thereto. It is understood that numerous variations not illustrated can be envisioned without departing from the scope of the invention.

Reference Signs List 10, 103e lower conductor pattern 111 second cold spray film 100 power semiconductor module 101 base plate 102 case 103 insulating substrate 103a, 103b, 103c upper conductor pattern 103d insulating material 103e lower conductor pattern , 104 semiconductor device, 104a switching device, 104b freewheeling diode, 105 sealing resin, 106 wiring, 106a plate wiring, 106b source signal wiring, 106c gate signal wiring, 107a, 107b, 107c solder, 108a main terminal, 108b control terminal , 109 upper surface electrode, 109a gate pad, 110 first cold spray film, 112 protective film, 113 third cold spray film, 201 semiconductor substrate, 201a first main surface, 201b second main surface, 202 Al alloy electrode, 204 barrier Metal, 205 Ni electrode, 206 Au electrode, 210 First main electrode, 220 Second main electrode, 301 Semiconductor wafer, 302 Semiconductor device formation region, 304 Dicing line, 404 Mask, 420 Cold spray device, 421 Spray gun, 422 Nozzle , 423 powder, 424 piping, 500 power supply, 510 power converter, 511 main conversion circuit, 512 semiconductor module, 513 control circuit, 520 load, OP opening.

Claims (9)

a semiconductor substrate;
a surface electrode provided on the first main surface of the semiconductor substrate;
a protective film provided on the periphery of the surface electrode;
a first metal film disposed on a region of the surface electrode surrounded by the protective film, separated from the protective film and in contact with the surface electrode ;
The semiconductor device according to claim 1, wherein the first metal film is a cold-spray film, and is joined to the plate-like wiring by solder spaced apart from the protective film. further comprising a second metal film provided on the peripheral edge of the first metal film between the first metal film and the plate-like wiring;
A semiconductor device according to claim 1 . The second metal film has better solder wettability than the first metal film,
3. The semiconductor device according to claim 2. wherein the second metal film is a cold spray film;
4. The semiconductor device according to claim 2 or 3. A semiconductor device according to any one of claims 1 to 4;
provided at a position facing the first metal film and joined to the first metal film with the solder,
Power semiconductor module. A through hole is formed at a position facing the first metal film of the plate-shaped wiring,
6. The power semiconductor module according to claim 5. A semiconductor device according to any one of claims 2 to 4;
The plate-shaped wiring is provided at a position facing the first metal film and is joined to the first metal film with the solder,
there is a gap between the second metal film and the plate-like wiring;
Power semiconductor module. A semiconductor device having the semiconductor device according to any one of claims 1 to 4 or the power semiconductor module according to any one of claims 5 to 7, converting input power to output a main conversion circuit that
a control circuit that outputs a control signal for controlling the main conversion circuit to the main conversion circuit;
Power converter. forming a surface electrode on the first main surface of the semiconductor substrate;
forming a protective film on the peripheral edge of the surface electrode;
forming a first metal film on a region of the surface electrode surrounded by the protective film by a cold spray method , separated from the protective film and in contact with the surface electrode ;
bonding the first metal film to a plate-like wiring by solder spaced apart from the protective film;
A method of manufacturing a power semiconductor module.

Images