JP4620566B2 - Semiconductor device and manufacturing method thereof - Google Patents

Description

  The present invention relates to a semiconductor device and a manufacturing method thereof, and more particularly to a semiconductor device in which a lead has a function of adjusting a center of gravity and a manufacturing method thereof.

In a semiconductor device, an extraction electrode is provided in order to extract an electric signal from a semiconductor element sealed inside with a resin. Various lead electrode structures have been proposed in consideration of the reliability and electrical characteristics of semiconductor devices. For example, in a power semiconductor device through which a large current flows, a structure has been proposed in which a plate-like lead is joined to the surface of a semiconductor element as an extraction electrode (see, for example, Patent Document 1).
JP 2000-114445 A

  However, in such a semiconductor device, it is necessary to fix the lead to the surface of the semiconductor element and to fix the electrode terminal to the end portion of the lead, and the lead fixed to the semiconductor element and an external lead frame that becomes the electrode terminal The positional relationship was limited. Further, in order to manufacture a semiconductor device having a complicated circuit configuration, when the lead is fixed to the surface of the semiconductor element, it is necessary to hold the lead in a predetermined position, and the manufacturing process is complicated.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a power semiconductor device and a method for manufacturing the power semiconductor device in which the manufacturing process is simplified by using a lead having a function of adjusting the center of gravity.

The present invention relates to a block-shaped heat sink having a surface and a back surface facing each other, a semiconductor element having a surface and a back surface facing each other fixed to the surface of the heat sink, and a junction region of the surface of the semiconductor element The lead is fixed to the lead by the bonding material and the electrode terminal attached to the lead. When the gravity center adjustment member is fixed to the lead and the direction of gravity is in the normal direction of the heat sink, the lead The center of gravity of the center of gravity adjustment member is on the semiconductor element.

  The present invention also includes a step of preparing a heat sink having a surface and a back surface facing each other, a step of fixing a semiconductor element having a surface and a back surface facing each other on the surface of the heat sink, The fixing process includes fixing the lead to the bonding area on the surface and attaching the electrode terminal to the lead. The fixing process arranges the solder material in the bonding area and makes the lead stand up so that the lead contacts the solder material. It is also a method for manufacturing a semiconductor device, which is a step of melting a solder material in a state and fixing a lead to a joining region.

  According to the present invention, since the lead can be fixed on the surface of the semiconductor element in a self-supporting state, the productivity and yield of the semiconductor device are improved, the degree of freedom in designing the semiconductor device is improved, and a complicated circuit is easily formed. Can be obtained.

Embodiment 1 FIG.
FIG. 1 is a cross-sectional view of a power semiconductor device according to a first embodiment of the present invention, the whole being represented by 100, and FIG. 2 is a perspective view of the vicinity of the IGBT 2 of the semiconductor device 100 (resin casing 8). Is excluded).

  The power semiconductor device 100 includes, for example, a Cu block 1 having a length of 50 mm, a width of 30 mm, and a thickness of 3 mm. An IGBT (insulated gate bipolar transistor) 2 having a thickness of 0.4 mm, a length and a width of 15 mm, for example, is joined to the surface of the Cu block 1 by a Sn-Ag-Cu solder 3 having a thickness of 0.15 mm, for example. ing. The Cu block 1 enhances the heat dissipation effect of the IGBT 2 and also serves as a back surface wiring.

  In the junction region on the surface of the IGBT 2, for example, a joint portion with the IGBT 2 is a rectangular shape having a length of 13 mm × width of 10 mm, and a Cu lead 4 having a thickness of 0.2 mm is Sn-Ag— with a thickness of 0.2 mm, for example. Bonded by a Cu-based solder 5. Since it is necessary to provide a sufficient insulation distance between the Cu lead 4 and the Cu block 1, the Cu lead 4 is bent in a direction perpendicular to the surface of the IGBT 2 at the bent portion 4B (see FIG. 3), and thus the IGBT 2 And a distance of about 1.5 mm from the surface.

  In addition, when the Cu lead 4 used for joining is made thinner, cracks of the solder 5 are less likely to occur when a temperature cycle is applied, and the reliability is improved.

  The Cu lead 4 is provided with a center-of-gravity adjustment portion 4A at the end located on the IGBT 2 so that the Cu lead 4 is self-supporting on the junction region of the IGBT 2 (it does not tilt but remains stationary). . As shown in FIG. 3, the center-of-gravity adjusting portion 4A is formed, for example, by bending an end portion of the Cu lead 4 in a bending process during lead molding.

  The center-of-gravity adjusting portion 4A is manufactured so that the Cu lead 4 does not fall by the rotational moment due to its own weight with 4B shown in FIG. 3 as a fulcrum. In other words, a straight line passing through the center of gravity of the Cu lead 4 intersects the bonding area perpendicularly. The bent portion 4B is an end portion in the pulling direction of the joint portion between the Cu lead 4 and the IGBT 2.

Electrode terminals 6A and 6B are fixed to the Cu block 1 and the Cu lead 4 with Ag paste or the like, respectively, and current is input / output from / to the outside of the semiconductor device. An electrode terminal 6C is connected to the gate electrode (not shown) of the IGBT 2 using an Al wire 7.
At the joint portion between the electrode terminal 6B and the Cu lead 4, the Cu lead 4 and the electrode terminal 6B are overlapped and joined so that the Cu lead 4 is closer to the heat sink 1 than the electrode terminal 6B.

The Cu block 1 or the like is sealed in a resin casing 8 made of an epoxy resin or the like mixed with a filler.
Note that an insulating sheet 9 having a higher amount of filler added than the resin casing 8 and having a high thermal conductivity is fixed to the back surface of the Cu lead 1.

Next, a method for manufacturing a semiconductor device will be briefly described.
First, the Cu block 1 is prepared. On the Cu block 1 heated to about 280 ° C., a plate-shaped solder preformed to have approximately the same size as the area of the IGBT 2 is placed. The following soldering process is performed by a tunnel furnace passing through a reducing atmosphere furnace.

  Next, the IGBT 2 is placed on the melted solder so that a collector electrode (not shown) corresponding to the second main electrode is connected to the Cu block 1.

Next, the preformed plate-like solder is placed on the emitter electrode (joining region: not shown) corresponding to the first main electrode of the IGBT 2 and melted. Further, the Cu lead 4 is formed on the melted solder. Is placed.
In addition, on the first main electrode (joining region) of the IGBT 2, a Ni film having a thickness of about 5 μm is formed in advance by, for example, an electroless plating method in order to improve wettability with solder. Is preferred.

  Here, as shown in the cross-sectional view of the Cu lead 4 in FIG. 3, the Cu lead 4 is placed on the IGBT 2 so that the Cu lead 4 is self-supporting on the IGBT 2 (it does not tilt even if it is not fixed). The center of gravity adjustment portion 4A is provided at the end portion located. As shown in FIG. 3, the center-of-gravity adjusting portion 4A is formed, for example, by bending an end portion of the Cu lead 4 in a bending process during lead molding.

  The center-of-gravity adjusting portion 4A is manufactured so that the Cu lead 4 does not fall by a rotating moment due to its own weight, with the bent portion 4B shown in FIG. 3 as a fulcrum. Reference numeral 4B denotes an end portion in the pulling direction of a portion where the Cu lead 4 and the IGBT 2 are joined.

  FIG. 4 is a perspective view (excluding the resin casing 8) in the vicinity of the IGBT 2 of the semiconductor device 100, represented as a whole by 120. In the structure of FIG. 4, the center-of-gravity adjustment portion 4A is formed from two bent portions. Depending on the position and weight of the required center of gravity, the structure of the center of gravity adjusting unit 4A shown in FIGS.

  As will be described later, since the Al 2 wiring is provided in the control wiring of the IGBT 2, the center of gravity adjusting portion 4A is formed so as not to overlap the control electrode.

  After this process, the whole is cooled to complete the soldering. The positioning of the Cu lead 4 with respect to the IGBT 2 is sufficiently performed by self-alignment by melting the preformed solder and applying a wetting force to the Cu lead 4.

  After the soldering of the Cu lead 4 to the IGBT 2 is completed, a lead frame (not shown) constituting the internal electrode and the external electrode is stacked from above, and the Cu block 1 and the Cu lead 4 are joined to the lead frame. For the bonding, a conductive adhesive such as an Ag paste that can be bonded at a temperature lower than the melting point of the solder is used.

  Finally, the Cu block 1 or the like is sealed by a transfer molding method to form the resin casing 8. Further, the lead frame is cut by a press die or the like and bent to form electrode terminals 6A, 6B, 6C. As a result, the electrode terminals 6A and 6B are fixed to the Cu block 1 and the Cu lead 4, respectively, and current is input / output from / to the outside of the semiconductor device. An electrode terminal 6C is connected to the gate electrode (not shown) of the IGBT 2 using, for example, an Al wire 7.

  Note that an insulating sheet 9 having a higher thermal conductivity and having a larger amount of filler added than the resin casing 8 is fixed to a portion of the Cu block 1 that contacts the outside of the casing.

  Through the above steps, the power semiconductor device 100 according to the present embodiment is completed.

  In the first embodiment, the IGBT is described as an example of the semiconductor element. However, a MOSFET that is a power transistor may be used instead.

  In addition, Sn—Ag solder is exemplified as the solder material, but other conductive materials such as Au—Si, Sn—Pb, etc., which are usually used for electrical joints, and Ag-containing epoxy resin, etc. An adhesive may be used. In particular, when a conductive adhesive is used, since the Cu lead 4 does not have a holding power as high as the solder wettability, the Cu lead 4 must be configured to be self-supporting as in the present embodiment. The process of fixing the Cu lead 4 on the IGBT 2 becomes difficult.

  Further, the Ni film formed by the plating method is exemplified as the metal film having good solder wettability, but a Ni / Au laminated film formed by the sputtering method can also be used. Thus, if the surface is a metal that causes solder to wet, such as Ni, Ag, Au, Pd, or Cu, it can be used as a metal film.

  In the power semiconductor device 100 according to the present embodiment, the Cu lead 4 has the center-of-gravity adjustment portion 4A and is self-supporting on the IGBT (semiconductor element) 2 (it does not tilt and remains stationary without being fixed). In the joining process of the Cu lead 4 on the IGBT (semiconductor element) 2, it is not necessary to fix the Cu lead 4 so as not to be inclined with a spacer or a weight.

  In particular, when a spacer is used to support the Cu lead 4, it takes time to increase and decrease the temperature because the spacer is also heated and cooled in the heating and cooling processes for soldering. It was. On the other hand, since no spacer is used in the present embodiment, the time can be reduced by about 30% including the time for attaching the spacer.

  In addition, since the gap between the Cu block 1 and the Cu lead 4 which is a heat sink is fixed at the height of the spacer, when the amount of solder is small, the Cu lead 4 is mounted in a floating state, and the soldering area May decrease or may become open without soldering. On the other hand, in this embodiment, since no spacer is used, good soldering is possible.

Further, since it is not necessary to use a spacer for supporting the Cu lead 4, the productivity in soldering is improved, and the amount of production about 1.5 times that of the conventional method can be achieved.
In addition, the bending accuracy of the Cu lead 4 is not 0.05 mm, which is regulated at the joint, but is 0.2 mm because it is governed by the insulation distance from the semiconductor element (IGBT2) and the heat sink (Cu block 1). As a result, the cost of the Cu lead 4 can be reduced by about 20%.

  In addition, the circuit configuration in the semiconductor device 100 is only a heat sink (Cu block 1), a semiconductor element (IGBT2), and a lead. By simply selecting a lead frame by combining soldered elements, Two-element parallel 2IN1 modules and six-element parallel 6IN1 modules used in the inverter circuit can be formed.

  Further, by using the structure and bonding process shown in the present embodiment, it is possible to easily form a complicated circuit simply by stacking and bonding the lead frame on the heat sink and the leads.

  Further, in the semiconductor device 100 according to the present embodiment, when there is a large amount of heat in a short time or when the Cu lead 4 itself generates heat, heat dissipation to the resin housing 8 is promoted, and reliability with little temperature change is achieved. A semiconductor device with a high level can be obtained.

Embodiment 2. FIG.
FIG. 5 is a cross-sectional view of the power semiconductor device according to the second embodiment of the present invention, indicated as a whole by 200. In FIG. 5, the same reference numerals as those in FIG. 1 indicate the same or corresponding parts.

  In the semiconductor device 200 according to the second embodiment, the end portion of the Cu lead 4 has an end portion protruding from the surface of the Cu block 1 by about 5 mm (projecting rightward in FIG. 5).

In the process of connecting the electrode terminals 6A-6C to the Cu lead 4, the lead frame is mounted from above the Cu lead 4, and for example, an ultrasonic wave is applied by applying a pressure of about 50 MPa to the joint portion from above and below. Connecting.
In the semiconductor device 200 according to the present embodiment, since the end of the Cu lead 4 protrudes from the end of the surface of the Cu block 1 by about 5 mm, pressure is easily applied to the joint between the Cu lead 4 and the electrode terminal 6B. Can be added.

Therefore, even when there is a large variation in the height of the lead or the height of the frame, the frame can be sufficiently deformed, and the Cu lead can be supported by a jig or the like. It becomes possible.
In addition, a suitable method can be selected from a plurality of joining methods such as resistance welding, resistance brazing, and ultrasonic joining that require a pressure of several MPa. For this reason, for example, spot welding is used as appropriate for a semiconductor device with a small operating current of 100 A or less, and resistance brazing that can increase the connection area is used as appropriate for a semiconductor with a large operating current exceeding 300 A. can do.

  As described above, in the semiconductor device 200 according to the present embodiment, since the options of the joint shape and the joining method are widened, the degree of freedom in frame design is improved, and a semiconductor device in which a large number of semiconductor elements are combined is formed. This makes it easy to produce a circuit.

Embodiment 3 FIG.
FIG. 6 is a cross-sectional view of the power semiconductor device according to the third embodiment of the present invention, the whole being represented by 300. FIG. 7 is a perspective view of the vicinity of the IGBT 2 of the power semiconductor device 300. 6 and 7, the same reference numerals as those in FIGS. 1 and 2 indicate the same or corresponding portions.

  In the semiconductor device 300, for example, a tungsten center of gravity adjusting member 10 having a length of 1 mm, a width of 4 mm, and a height of 8 mm is formed on the surface of the Cu lead 4 facing the surface to be bonded to the emitter electrode (not shown) on the IGBT 2. For example, it is fixed in advance by Ag brazing or the like. By having such a center-of-gravity adjusting member 10, the Cu lead 4 is self-supporting.

  Here, the center-of-gravity adjusting member 10 may be fixed to the entire joint surface on the Cu lead 4, but tungsten or the like is an expensive material, and therefore, tungsten having a minimum necessary weight for the lead to stand on its own is used. It is preferable from the viewpoint of manufacturing cost reduction.

  In order to realize the self-supporting of the lead with smaller parts, it is preferable to use a material having a specific gravity equal to or higher than that of Cu, such as molybdenum or the above-described tungsten. The center-of-gravity adjusting member 10 is not limited to a conductive material, and may be an insulator.

  In the above first to third embodiments, description has been made by taking Cu as an example of the lead material. However, by using Al having a small specific gravity, the center-of-gravity adjusting member 10 can be reduced in size and thickness.

  For example, FIG. 8 shows a structure in which the current density and heat generation of the IGBT 2 are dispersed by dividing the wiring of the Cu lead 4 from the IGBT 2 that generates a large amount of heat into two directions. In the case of such a lead structure, it is extremely difficult to form a portion that adjusts the center of gravity of the lead by bending the lead because the height of the entire lead is more than three times the bending height. However, when the gravity center adjusting member 10 is attached, the attachment position can be selected relatively freely. For example, as shown in FIG. 8, the center-of-gravity adjusting member 10 may be installed in a region facing the joint portion, or may be installed in the wiring portion as shown in FIG.

  As described above, in the semiconductor device 300 according to the third embodiment, it is possible to manufacture a thin package semiconductor device in which the center of gravity position of the Cu lead cannot be adjusted only by deformation of the Cu lead.

Embodiment 4 FIG.
FIG. 10 is a perspective view of the vicinity of the IGBT 2 of the power semiconductor device according to the fourth embodiment. 10, the same reference numerals as those in FIG. 1 denote the same or corresponding parts.

  First, a solder sheet in which Cu leads 4 having both ends developed (both ends extended) so as to be self-supporting on the joining region of the Cu block 1, IGBT 2 and IGBT 2 are preformed in accordance with the shape of each joining region. (Not shown). As a soldering method, the laminated soldering using a tunnel furnace as shown in the first embodiment is used.

  At this time, a notch 4C having a length of 2 mm and a width of 2 mm is provided on the side surface of the Cu lead 4 covering the control electrode of the IGBT 2.

  After the soldering is completed, as shown in FIG. 11, the end of the Cu lead 4 covering the control electrode of the Cu lead 4 is bent toward the opposite end, and the overlapping Cu leads 4 are For example, it is fixed by ultrasonic bonding.

  Thereafter, the lead frame (not shown) constituting the internal electrode and the external electrode is overlapped from above, and the lead frame and the Cu lead 4 are electrically connected with a conductive adhesive such as an Ag paste that can be connected at a solder melting point or lower. Connect.

  In the power semiconductor device manufactured by such a manufacturing method, it has been shown in the first embodiment that the spacers and weights are not required by using the Cu leads 4 that are self-supported in the unfolded state. It is the same as that. Here, since the Cu lead 4 is further overlapped and joined, the cross-sectional area of the current path from the IGBT 2 or the like can be approximately doubled.

  When the Cu lead 4 is thinned to improve the temperature cycle reliability on the IGBT 2, the electric resistance of the wiring portion increases. In contrast, the semiconductor device according to the present embodiment can improve the reliability and produce a semiconductor device with low wiring resistance.

It is sectional drawing of the semiconductor device for electric power concerning Embodiment 1 of this invention. 1 is a partial perspective view of a power semiconductor device according to a first embodiment of the present invention. It is sectional drawing of Cu lead used for the semiconductor device for electric power concerning Embodiment 1 of this invention. It is a fragmentary perspective view of the other power semiconductor device concerning Embodiment 1 of this invention. It is sectional drawing of the semiconductor device for electric power concerning Embodiment 2 of this invention. It is sectional drawing of the semiconductor device for electric power concerning Embodiment 3 of this invention. It is a fragmentary perspective view of the semiconductor device for electric power concerning Embodiment 3 of this invention. It is a fragmentary perspective view of the other power semiconductor device concerning Embodiment 3 of this invention. It is a fragmentary perspective view of the other power semiconductor device concerning Embodiment 3 of this invention. It is a fragmentary perspective view of the semiconductor device for electric power concerning Embodiment 4 of this invention. It is a fragmentary perspective view of the semiconductor device for electric power concerning Embodiment 4 of this invention.

Explanation of symbols

  1 Cu block, 2 IGBT, 3 solder, 4 Cu lead, 4A center of gravity adjustment part, 4B fulcrum, 4C notch, 5 solder, 6A, 6B, 6C electrode terminal, 7 Al wire, 8 resin housing, 9 resin sheet, 10 Center of gravity adjustment member, 100 semiconductor device.

Claims (6)

A block-shaped heat sink having a front surface and a back surface facing each other;
A semiconductor element having a front surface and a back surface that are opposed to each other and fixed to the surface of the heat sink;
A lead fixed with a bonding material to a bonding region on the surface of the semiconductor element;
A semiconductor device including an electrode terminal attached to the lead,
A semiconductor device, wherein a center of gravity of the lead and the center of gravity adjusting member is on a semiconductor element when the center of gravity adjusting member is fixed to the lead and the direction of gravity is in the normal direction of the back surface of the heat sink. A block-shaped heat sink having a front surface and a back surface facing each other;
A semiconductor element having a front surface and a back surface that are opposed to each other and fixed to the surface of the heat sink;
A lead fixed with a bonding material to a bonding region on the surface of the semiconductor element;
A semiconductor device including an electrode terminal attached to the lead,
The lead should be bent so that one of the leads extending in the opposite direction is connected to the other so that the center of gravity of the lead is on the semiconductor element when the direction of gravity is in the normal direction of the back surface of the heat sink. A semiconductor device characterized by the above.   3. The lead and the electrode terminal according to claim 1 or 2, wherein the lead and the electrode terminal are overlapped so that the lead is closer to the heat sink than the electrode terminal at a joint portion between the electrode terminal and the lead. The semiconductor device described.   The lead extends substantially parallel to the surface of the heat sink, has an outer end extending outward beyond the end of the heat sink, and the electrode terminal is attached to the outer end. The semiconductor device according to claim 1, wherein: Preparing a heat sink with front and back surfaces facing each other;
Fixing a semiconductor element having a front surface and a back surface facing each other on the surface of the heat sink;
A fixing step of fixing the lead to the bonding region on the surface of the semiconductor element;
Attaching the electrode terminal to the lead,
The fixing step is a step of disposing the solder material in the joining region, melting the solder material in a state where the lead is self-supported so that the lead is in contact with the solder material, and fixing the lead to the joining region. A method for manufacturing a semiconductor device, comprising:   6. The method of manufacturing a semiconductor device according to claim 5, further comprising a step of bending so that one of the leads extending in opposite directions across the joining region is connected to the other after the fixing step.

Images