Semiconductor device manufacturing method

Abstract

To provide a manufacturing method for a semiconductor device, in which an electrode member can be extracted for sure from an upper surface of a device while suppressing the manufacturing cost.SOLUTION: A semiconductor chip 1 including a main electrode 5 and a control electrode 6 is bonded to a substrate 3. A wiring chip 2 including a first electrode 8, a second electrode 9, and a wire 10 connecting the first electrode 8 and the second electrode 9 is bonded to the substrate 3. A main electrode member 12 is bonded to a main electrode 5 through a first bonding material 13. The control electrode 6 and the first electrode 8 are connected to each other with a connection member 16. These are put in a mold 18, and with a tip end surface of the main electrode member 12 pressed against a buffer material 19 provided between the main electrode member 12 and the mold 18, a sealing material 17 is injected into the mold 18 and sealed. A tip end part of the main electrode member 12 protrudes from an upper surface of the sealing material 17. The sealing material 17 provided at a side surface of the tip end part has a tapered shape whose thickness decreases toward the tip end surface.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present disclosure relates to a method for manufacturing a semiconductor device. [Background technology]

[0002] Because it is difficult to increase the area of ​​SiC MOSFETs, modules using SiC MOSFETs require multiple chips to be connected in parallel to increase current capacity. A semiconductor device has been proposed in which multiple semiconductor chips and a wiring chip are bonded to a substrate, and the control electrodes of each semiconductor chip are connected in parallel using the circuit pattern of the wiring chip (see, for example, Patent Document 1). A main electrode member is bonded to the main electrodes of the multiple semiconductor chips, and a control electrode member is bonded to the circuit pattern of the wiring chip, and then the resulting product is sealed with resin. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2020 / 110170 Summary of the Invention [Problem to be solved by the invention]

[0004] In conventional technology, the sealing material is ground to expose the electrode members. However, dust is generated during the grinding process, and in the case of wet grinding, moisture can penetrate through the ground surface. Furthermore, variations in the height of the electrode members can result in more electrode members being ground than necessary during grinding, resulting in wasted material and processing time, and making it difficult to detect the end point. Furthermore, the grinding process itself is costly.

[0005] The present disclosure has been made to solve the above-mentioned problems, and its purpose is to provide a method for manufacturing a semiconductor device that can reliably remove electrode members from the top surface of the device while reducing manufacturing costs. [Means for solving the problem]

[0006] a step of bonding a semiconductor chip having a main electrode and a control electrode to a substrate; a step of bonding a wiring chip having a first electrode, a second electrode, and wiring connecting the first electrode and the second electrode to the substrate; a step of bonding a main electrode member to the main electrode via a bonding material; a step of connecting the control electrode and the first electrode with a connecting material; and a step of placing the bonded semiconductor chip, the substrate, the wiring chip, the main electrode member, and the connecting material in a mold, and injecting a sealing material into the mold while pressing a tip end surface of the main electrode member against a buffer material provided between the main electrode member and the mold, to seal the semiconductor chip, the substrate, the wiring chip, the main electrode member, and the connecting material with the sealing material, wherein a tip end of the main electrode member protrudes from an upper surface of the sealing material, and the sealing material provided on a side surface of the tip end has a tapered shape whose thickness becomes thinner toward the tip end surface. and does not grind the sealing material. It is characterized by: [Effects of the Invention]

[0007] In this disclosure, even if a gap occurs between the main electrode member and the mold due to variations in the height of the main electrode member, the buffer material fills the gap. Therefore, the leading edge of the main electrode member is exposed and not covered by the sealing material during the sealing process, so the sealing material does not need to be ground after the sealing process. This allows the electrode member to be reliably removed from the top surface of the device while reducing manufacturing costs. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a cross-sectional view showing a semiconductor device according to a first embodiment. [Figure 2] 2A to 2C are cross-sectional views showing a method for manufacturing the semiconductor device according to the first embodiment. [Figure 3] 2A to 2C are cross-sectional views showing a method for manufacturing the semiconductor device according to the first embodiment. [Figure 4] 2A to 2C are cross-sectional views showing a method for manufacturing the semiconductor device according to the first embodiment. [Figure 5] 2A to 2C are cross-sectional views showing a method for manufacturing the semiconductor device according to the first embodiment. [Figure 6] 2A to 2C are cross-sectional views showing a method for manufacturing the semiconductor device according to the first embodiment. [Figure 7] 2A to 2C are cross-sectional views showing a method for manufacturing the semiconductor device according to the first embodiment. [Figure 8] 10A to 10C are cross-sectional views showing a method for manufacturing a semiconductor device according to a comparative example. [Figure 9] 10A to 10C are cross-sectional views showing a method for manufacturing a semiconductor device according to a comparative example. [Figure 10] 10A to 10C are cross-sectional views showing a method for manufacturing a semiconductor device according to a comparative example. [Figure 11] FIG. 4 is a cross-sectional view showing the tip of the main electrode member after sealing. [Figure 12] 10A and 10B are cross-sectional views showing a modified example of the method for manufacturing the semiconductor device according to the first embodiment. [Figure 13] FIG. 10 is a cross-sectional view showing a semiconductor device according to a second embodiment. [Figure 14] 10A to 10C are cross-sectional views showing a method for manufacturing a semiconductor device according to a second embodiment. [Figure 15] 10A to 10C are cross-sectional views showing a method for manufacturing a semiconductor device according to a second embodiment. [Figure 16] 10A to 10C are cross-sectional views showing a method for manufacturing a semiconductor device according to a second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] A method for manufacturing a semiconductor device according to an embodiment will be described with reference to the drawings. The same or corresponding components are designated by the same reference numerals, and repeated description may be omitted.

[0010] Embodiment 1 FIG. 1 is a cross-sectional view showing a semiconductor device according to a first embodiment. A plurality of semiconductor chips 1 and one wiring chip 2 are bonded to a substrate 3. The semiconductor chip 1 has a back electrode 4 on its back side and a main electrode 5 and a control electrode 6 on its front side. The semiconductor chip 1 is, for example, a MOSFET. The control electrode 6 is, for example, a gate electrode or a Kelvin source electrode. If the semiconductor chip 1 has a temperature sensing element or a current sensing element built in, it will further be provided with a control electrode corresponding to each of them.

[0011] The back electrode 4 is, for example, a metal film formed by laminating a silicide layer / titanium layer / nickel layer / titanium layer / gold or silver layer from the semiconductor chip 1 side by a sputtering method. The main electrode 5 and the control electrode 6 are, for example, metal films formed by forming an aluminum layer on the semiconductor chip 1 by a sputtering method and laminating a nickel layer / palladium layer / gold layer thereon by a plating method. Alternatively, the main electrode 5 and the control electrode 6 may be, for example, metal films formed by laminating an aluminum layer / titanium layer / nickel layer / titanium layer / gold or silver layer by a sputtering method. Other laminated films having similar functions may also be selected.

[0012] The wiring chip 2 has a bonding layer 7 on its back surface side and a first electrode 8, a second electrode 9, and wiring 10 connecting the first electrode 8 and the second electrode 9 on its front surface side. The wiring chip 2 is an element made of, for example, silicon, with an insulating film such as an oxide film formed on the Si surface, and the first electrode 8, the second electrode 9, and wiring 10 formed on the insulating film. The first electrode 8, the second electrode 9, and wiring 10 are wiring patterns made of, for example, aluminum. A bondable metal layer is provided at least on the second electrode 9. A similar metal layer may also be provided on the first electrode 8. The bondable metal layer is, for example, a laminated metal film similar to the main electrode 5 of the semiconductor chip 1. The bonding layer 7 is, for example, a metal film similar to the back electrode 4 of the semiconductor chip. The back electrode 4 of the semiconductor chip 1 and the bonding layer 7 of the wiring chip 2 are each bonded to the substrate 3 via a bonding material 11.

[0013] A main electrode member 12 is bonded to the main electrode 5 of the semiconductor chip 1 via a first bonding material 13. The main electrode member 12 is made of, for example, copper. When the main electrode member 12 spans the main electrodes 5 of multiple semiconductor chips 1, the portion that bonds with the main electrodes 5 protrudes, and the portion connecting them is thinner than the portion that bonds with the main electrodes 5. This makes it possible to connect the main electrodes 5 of multiple semiconductor chips 1 to the same potential, avoiding the peripheral voltage-resistant structure of the semiconductor chip 1. Alternatively, the main electrodes 5 of multiple semiconductor chips 1 may be independent of each other and have the same potential when connected to an external electrode.

[0014] A control electrode member 14 is bonded to the second electrode 9 of the wiring chip 2 via a second bonding material 15. The control electrode member 14 is a plurality of blocks made of copper, for example.

[0015] The control electrodes 6 of the multiple semiconductor chips 1 and the first electrodes 8 of the wiring chip 2 are connected by connecting members 16. The connecting members 16 are wires made of, for example, gold, silver, or aluminum. Using thin wires made of gold or silver allows the dimensions of the control electrodes 6 of the semiconductor chip 1 to be reduced. This allows the effective area to be increased, and the manufacturing cost of the semiconductor chip 1 to be reduced.

[0016] The semiconductor chip 1, the upper surface of the substrate 3, the wiring chip 2, the main electrode member 12, the control electrode member 14, and the connection member 16 are sealed with a sealing material 17. The sealing material 17 is, for example, an epoxy resin mixed with a filler. The tips of the main electrode member 12 and the control electrode member 14 protrude from the upper surface of the sealing material 17, and their tip faces are exposed from the sealing material 17.

[0017] Next, a method for manufacturing a semiconductor device according to this embodiment will be described. FIGS. 2 to 7 are cross-sectional views illustrating the method for manufacturing a semiconductor device according to the first embodiment. First, as shown in FIG. 2, a semiconductor chip 1 is bonded to a substrate 3. Next, as shown in FIG. 3, a wiring chip 2 is bonded to the substrate 3. At this time, the back electrode 4 of the semiconductor chip 1 and the bonding layer 7 of the wiring chip 2 are each bonded to the substrate 3 via a bonding material 11. The bonding material 11 may be, for example, solder, or a bonding material made of silver or copper may be used for sinter bonding. The sinter bonding may be pressure bonding, in which the semiconductor chip 1 and the wiring chip 2 are heated while applying pressure to the electrodes from above, or pressureless bonding, in which no pressure is applied. Alternatively, the wiring chip 2 may be bonded using an adhesive or the like that provides thermal and mechanical stability by heating. However, when using a wiring chip 2 made of silicon, a bonding method that does not involve pressure is preferable because the wiring chip 2 may crack.

[0018] Next, as shown in Fig. 4, the main electrode member 12 is joined to the main electrode 5 via a first joining material 13. Next, as shown in Fig. 5, the control electrode member 14 is joined to the second electrode 9 via a second joining material 15. The first joining material 13 and the second joining material 15 may be, for example, solder, or may be joined by sinter bonding using a joining material made of silver or copper. Sinter bonding may be pressure bonding, in which the electrodes are heated while being pressed down from above, but pressureless bonding, in which no pressure is applied, is preferable.

[0019] The thickness of the main electrode member 12 and the control electrode member 14 varies due to manufacturing tolerances. Also, the thickness of the first bonding material 13 and the second bonding material 15 varies. Therefore, the height of the tip surface of the joined main electrode member 12 and the tip surface of the joined control electrode member 14 differs.

[0020] Next, as shown in Fig. 6, the control electrode 6 and the first electrode 8 are connected by a connecting member 16. Next, as shown in Fig. 7, the semi-finished product having the joined semiconductor chip 1, substrate 3, wiring chip 2, main electrode member 12, control electrode member 14, and connecting member 16 is placed into a mold 18. At this time, a buffer material 19 is provided between the mold 18 and the main electrode member 12 and control electrode member 14.

[0021] When the tip surfaces of the main electrode member 12 and the control electrode member 14 are pressed against the cushioning material 19, the thickness of the cushioning material 19 at the contact points between the main electrode member 12 and the control electrode member 14 becomes thinner due to the pressing force. If the tip surfaces of the main electrode member 12 and the control electrode member 14 are at different heights, the thickness of the cushioning material 19 will differ at the contact points with the main electrode member 12 and the contact points with the control electrode member 14. In this state, the transfer mold sealing method is used to inject sealing material 17 into a mold 18, and the semiconductor chip 1, the upper surface of the substrate 3, the wiring chip 2, the main electrode member 12, the control electrode member 14, and the connection member 16 are sealed with sealing material 17.

[0022] When the semiconductor device is removed from the mold 18 after sealing, the buffer material 19 is also removed from the top surface of the semiconductor device. Through the above steps, a semiconductor device is manufactured in which the tip surfaces of the main electrode member 12 and the control electrode member 14 are exposed from the sealing material 17 without having to grind the sealing material 17.

[0023] Furthermore, by using multiple manufactured semiconductor devices, it is possible to construct a higher-level semiconductor device such as a half-bridge circuit or a full-bridge circuit. In this case, the substrate 3 is electrically and thermally connected to the drain circuit pattern by solder bonding or sinter bonding. The main electrode member 12 is electrically connected to the source circuit pattern by wire bonding, ribbon bonding, or lead frame solder bonding. The periphery of the semiconductor device and the circuit pattern are then covered with a secondary sealing material such as gel to manufacture the higher-level semiconductor device. Note that when solder is used as a bonding material for the semiconductor chip 1, it is desirable to use a high-melting-point solder with a melting point higher than the bonding process temperature when assembling the higher-level semiconductor device.

[0024] Next, the effects of this embodiment will be described in comparison with a comparative example. Figures 8 to 10 are cross-sectional views showing a method for manufacturing a semiconductor device according to the comparative example. In the comparative example, as shown in Figure 8, no buffer material 19 is used in the resin sealing process. This causes gaps between the mold 18 and the main electrode member 12 and the control electrode member 14, and the sealing material 17 gets in between. As a result, as shown in Figure 9, there are cases where the main electrode member 12 or the control electrode member 14 cannot be exposed from the surface of the sealing material 17. For this reason, as shown in Figure 10, it is necessary to grind away the excess sealing material 17 and parts of the main electrode member 12 and the control electrode member 14 to expose the main electrode member 12 and the control electrode member 14.

[0025] In contrast, in this embodiment, even if the heights of the control electrode member 14 and the main electrode member 12 vary and gaps occur between the main electrode member 12 and the mold 18, the buffer material 19 fills the gaps. Therefore, the tip surfaces of the main electrode member 12 and the control electrode member 14 are exposed and not covered with the sealant 17 during the sealing process, so the sealant 17 does not need to be ground after the sealing process. This allows the electrode members to be reliably removed from the top surface of the device while keeping manufacturing costs down.

[0026] 11 is a cross-sectional view showing the tip of the main electrode member after sealing. When the tip surfaces of the main electrode member 12 and the control electrode member 14 are pressed against the buffer material 19, the portion of the buffer material 19 that contacts the control electrode member 14 and the main electrode member 12 becomes thinner than the portion that does not contact. The buffer material 19 gradually becomes thinner from the portion that does not contact toward the portion that contacts. Since resin sealing is performed in this state, the shape of the sealant 17 follows the shape of the buffer material 19. Therefore, the sealant 17 provided on the side surfaces of the tip of the main electrode member 12 and the control electrode member 14 has a tapered shape in which the film thickness in the horizontal direction of the drawing becomes thinner toward the tip surfaces of the main electrode member 12 and the control electrode member 14.

[0027] If there is a difference in the linear expansion coefficients of the main electrode member 12 and the control electrode member 14 and the sealing material 17, stress will be generated between them due to thermal cycling or the like. If the stress causes cracks in the sealing material 17 or causes the sealing material 17 to peel off, moisture will penetrate during a moisture resistance test or the like and reach the semiconductor chip 1, shortening the life of the semiconductor device. In contrast, in this embodiment, as described above, the thickness of the sealing material 17 gradually decreases toward the tip surfaces of the main electrode member 12 and the control electrode member 14. Therefore, the stress decreases toward the end surfaces of the main electrode member 12 and the control electrode member 14, preventing the sealing material 17 from peeling off from the control electrode member 14 and the main electrode member 12.

[0028] Furthermore, if the semiconductor chip 1 has a built-in temperature sensing element or current sensing element, these elements are often more susceptible to static electricity than the main portion. Therefore, it is preferable that the surface of the buffer material 19, which comes into contact with the tip surfaces of the main electrode member 12 and the control electrode member 14, is conductive. This prevents a potential difference from occurring between the terminals during and after sealing, thereby preventing overvoltage breakdown of the semiconductor chip 1 due to static electricity.

[0029] FIG. 12 is a cross-sectional view illustrating a modified example of the manufacturing method for the semiconductor device according to the first embodiment. The buffer material 19 includes a non-conductive sealing material 17 and a conductive thin film 20, such as a carbon sheet or metal foil, disposed between the sealing material 17 and the main electrode member 12 and the control electrode member 14. The leading edges of the main electrode member 12 and the control electrode member 14 contact the conductive thin film 20, eliminating potential differences between the terminals and preventing overvoltage breakdown of the semiconductor chip 1 due to static electricity. Furthermore, the use of the conductive thin film 20 allows for the use of an inexpensive sealing material 17, such as Teflon (registered trademark). Alternatively, the buffer material 19 may be primarily composed of carbon. Because such a buffer material 19 is conductive, it can be constructed from a single material without the conductive thin film 20. This reduces processing costs.

[0030] Embodiment 2 13 is a cross-sectional view showing a semiconductor device according to embodiment 2. In this embodiment, there are no wiring chips 2 and no connecting members 16, and a control electrode member 14 is bonded to a control electrode 6 of a semiconductor chip 1 via a second bonding material 15. The other configurations are the same as those of embodiment 1.

[0031] Next, a method for manufacturing a semiconductor device according to this embodiment will be described. Figures 14 to 16 are cross-sectional views showing a method for manufacturing a semiconductor device according to embodiment 2. First, as shown in Figure 14, a semiconductor chip 1 is bonded to a substrate 3. Next, as shown in Figure 15, a main electrode member 12 is bonded to a main electrode 5 via a first bonding material 13. A control electrode member 14 is bonded to a control electrode 6 via a second bonding material 15.

[0032] 16, the semi-finished product having the bonded semiconductor chip 1, substrate 3, main electrode member 12, and control electrode member 14 is placed in a mold 18. At this time, buffer material 19 is provided between the mold 18 and the main electrode member 12 and control electrode member 14. With the tip surfaces of the main electrode member 12 and control electrode member 14 pressed against the buffer material 19, a sealant 17 is injected into the mold 18, and the semiconductor chip 1, substrate 3, main electrode member 12, and control electrode member 14 are sealed with the sealant 17.

[0033] When the semiconductor device is removed from the mold 18 after encapsulation, the buffer material 19 is also removed from the top surface of the semiconductor device. Through the above steps, a semiconductor device is manufactured in which the tip surfaces of the main electrode member 12 and the control electrode member 14 are exposed from the encapsulant 17 without having to grind the encapsulant 17. As a result, as with the first embodiment, the electrode members can be reliably removed from the top surface of the device while keeping manufacturing costs down. When the number of semiconductor chips connected in parallel is small, such as in small-capacity semiconductor products, the present embodiment, in which electrodes are individually taken out from the semiconductor chip 1 and connected to an external circuit without using the wiring chip 2, may be preferable because it keeps manufacturing costs down.

[0034] The semiconductor chip 1 is not limited to those made of silicon, but may also be made of a wide-bandgap semiconductor with a wider bandgap than silicon. Examples of wide-bandgap semiconductors include silicon carbide, gallium nitride-based materials, and diamond. Semiconductor chips made of such wide-bandgap semiconductors have high voltage resistance and allowable current density, allowing for miniaturization. By using this miniaturized semiconductor chip, semiconductor devices incorporating the semiconductor chip can also be miniaturized and highly integrated. Furthermore, the high heat resistance of the semiconductor chip allows for miniaturization of the heat dissipation fins of the heat sink, enabling the water-cooled section to be replaced by air-cooled, further miniaturizing the semiconductor device. Furthermore, the low power loss and high efficiency of the semiconductor chip allow for high efficiency in the semiconductor device. [Explanation of symbols]

[0035] REFERENCE SIGNS LIST 1 semiconductor chip, 2 wiring chip, 3 substrate, 5 main electrode, 6 control electrode, 8 first electrode, 9 second electrode, 10 wiring, 13 first bonding material, 15 second bonding material, 16 connection member, 17 sealing material, 18 mold, 19 buffer material, 20 conductive thin film

Claims

1. bonding a semiconductor chip having a main electrode and a control electrode to a substrate; a step of bonding a wiring chip having a first electrode, a second electrode, and wiring connecting the first electrode and the second electrode to the substrate; a step of joining a main electrode member to the main electrode via a joining material; connecting the control electrode and the first electrode with a connecting member; and placing the joined semiconductor chip, the substrate, the wiring chip, the main electrode member, and the connection member into a mold, and injecting a sealing material into the mold while pressing a tip end surface of the main electrode member against a buffer material provided between the main electrode member and the mold, thereby sealing the semiconductor chip, the substrate, the wiring chip, the main electrode member, and the connection member with the sealing material. a tip end of the main electrode member protrudes from an upper surface of the sealing material; the sealing material provided on the side surface of the tip portion has a tapered shape in which the film thickness becomes thinner toward the tip surface, A method for manufacturing a semiconductor device, wherein the sealing material is not ground.

2. bonding a semiconductor chip having a main electrode and a control electrode to a substrate; a step of joining a main electrode member to the main electrode via a joining material; and placing the joined semiconductor chip, the substrate, and the main electrode member in a mold, and injecting a sealing material into the mold while pressing a tip end surface of the main electrode member against a buffer material provided between the main electrode member and the mold, thereby sealing the semiconductor chip, the substrate, and the main electrode member with the sealing material. a tip end of the main electrode member protrudes from an upper surface of the sealing material; the sealing material provided on the side surface of the tip portion has a tapered shape in which the film thickness becomes thinner toward the tip surface, A method for manufacturing a semiconductor device, wherein the sealing material is not ground.

Images