Semiconductor device and a method of manufacturing the same

Abstract

To improve the connection reliability at the time of packaging a semiconductor device and to make the method management easy in a semiconductor device manufacturing method.The semiconductor device comprises: a tub 1e for supporting a semiconductor chip 2; a sealing body 3 formed by sealing the semiconductor chip 2 with a resin; a plurality of leads 1a made of a copper alloy, exposed to the back face 3a of the sealing body 3, and having a soldered layer 8 on the exposed mounted face 1d; and wires 4 for connecting the pads 2a of the semiconductor chip 2 and the corresponding leads 1a. In the manufacture method, the sealing body 3 is polished, after resin-molded, at its back face 3a with a brush to form the two widthwise edge portions, as exposed from the back face 3a of the sealing body 3, of the lead la into rounded faces, and the mounted face 1d of the lead la including the rounded faces is protruded at its central portion from the back face 3a of the sealing body 3 thereby to improve the connection reliability at the packaging time.

Description

FIELD OF THE INVENTION

The present invention relates to a semiconductor manufacturing technique and, more particularly, to a technique which is effective when applied to an improvement in the connection reliability at the time of mounting a surface mount type semiconductor device.

DESCRIPTION OF THE RELATED ART

The technique to be explained has been examined by us when we investigated and completed the invention, as will be summarized in the following.

One resin-sealed type semiconductor device, as intended to reduce the size, is exemplified by a semiconductor package called the “QFN” (Quad Flat Non-leaded Package).

The QFN is a surface mount in which a plurality of thin leads arranged around a tub supporting a semiconductor chip are arranged on the back face (i.e., the face on the semiconductor device mounting side) of a sealing body, and has a single-face molding structure which uses a lead frame having only one resin-molded face.

In the QFN, the individual leads are arranged on the back face of the sealing body so that a packaging area (of the connecting region) has been retained for mounting the leads with a solder on the mount substrate.

Therefore, the QFN is molded with a film sheet so that burrs (such as resin burr of a leakage resin or resin flush, as will be called the “resin burr”) of a molding resin may not be formed on the leads on the back face of the sealing body (although this molding method will be called the “laminate molding method”).

At the molding time, more specifically, the film sheet is held in close contact with the back face of the sealing body in the cavity of the molding die, so that the resin burr may be prevented from sticking to the leads on the back face of the sealing body.

Here, the structure of the QFN is described, for example, on pages 56 and 57 of “Semiconductor Assembling/Testing Technique of '99, Extra Issue of 1998 of Monthly Semiconductor World 1998”, issued on Jul. 27, 1998 by Kabushiki Gaisha PRESS JOURNAL”.

SUMMARY OF THE INVENTION

In the QFN of the technique thus far described, however, the stand-off height at the packaging time is made substantially as small as the plated thickness by its structure. Therefore, the solder wettability at the QFN packaging time is seriously deteriorated if the resin burr is formed in the vicinity of the leads on the back face of the sealing body.

In the QFN, moreover, leads la may be deformed toward the side of a semiconductor chip 2 , as exemplified in the comparison of FIG. 19A , by the influence such as the resin charging pressure at the molding time. As in a QFN 25 of the comparison of 19 B, a resin burr 16 such as the resin flash is then easily formed on the lower side (or back side) of the deformed lead la.

This resin burr 16 raises a problem that it can hardly be removed by the well-known deburring method such as the high-pressure water method or the liquid honing method.

In the semiconductor package having the single-face molding structure such as the QFN, on the other hand, the laminate molding method may be performed. Even with this laminate molding method, however, it is difficult to eliminate the resin burr completely. Another problem is that it takes a cost to make a modification into the laminate molding apparatus.

Where the QFN is molded by the laminate molding method, moreover, the molding resin charged between the leads may be wrinkled to form protrusions of the film sheet. These protrusions obstruct the lead clamping in the cutting die at the lead cutting time thereby to cause a problem that the leads are cut to become defective.

Here at the lead cutting step, the leads are pressed and cut downward (in the direction of the face on the side opposed to the mounted face) by the cutting punch (or cutting blade). Then, the lead burr is formed on the lower side (or on the mounted side) of the leads to cause the defective connection at the QFN packaging time thereby to raise a problem that the connection reliability is lowered.

On the other hand, the QFN is troubled by another problem that the materials of the individual components are deformed by the heat at the working time so that they cannot retain their flatnesses.

In the manufacture of the QFN, there is another problem that the facial pressure on the molding die has to be strictly managed so as to prevent the back face of the sealing body from having the resin burr.

At this time, another problem is that the molding die has to be frequently cleaned.

On the other hand, the technique for exposing the leads to the back face of the semiconductor device by polishing the back side is described in Japanese Patent Laid-Open No. 195743/99 and 240940/90. However, the former has no description on the semiconductor device having the single-face molding structure. The latter is directed to the technique for separating an integral lead into individual leads by polishing the back face side of the semiconductor device, but does not have any description on the semiconductor device in the state where the inner portions of the adjoining leads are separated in advance.

An object of the invention is to provide a semiconductor device for improving the connection reliability at a packaging time, and a method of manufacturing the same.

Another object of the invention is to provide a semiconductor device intended to facilitate the process management, and a method of manufacturing the same.

The aforementioned and other objects and novel features of the invention will become apparent from the following description to be made with reference to the accompanying drawings.

The representative of the invention disclosed herein will be briefly described in the following.

According to an aspect of the invention, there is provided a resin-sealed type semiconductor device comprising: a tub for supporting a semiconductor chip; a sealing body formed by sealing the semiconductor chip with a resin; a plurality of leads arranged around the tub and exposed to the face of the sealing body on the semiconductor device mounting side; and connecting members for connecting the surface electrodes of the semiconductor chip and the corresponding ones of the leads. The two widthwise edge portions of each of the leads are formed of rounded faces, and the mounted face of the lead including the rounded faces is protruded at its central portion from the face of the sealing body on the semiconductor device mounting side.

According to another aspect of the invention, there is provided a method of manufacturing a resin-sealed type semiconductor device, comprising the steps of: preparing a lead frame in which a plurality of leads are arranged around a tub for supporting a semiconductor chip; bonding the tub and the semiconductor chip; connecting the surface electrodes of the semiconductor chip and the corresponding ones of the leads by connecting members; forming a sealing body by resin-sealing the semiconductor chip and the connecting members on the lead frame on the semiconductor chip arranging side, to arrange the plurality of leads on the face of the semiconductor device mounting side; exposing the plurality of leads by polishing the face of the sealing body on the semiconductor device mounting side; and separating the plurality of leads from the frame portion of the lead frame.

According to the invention, the connection regions of the leads needed at the time of mounting the semiconductor device on the mount substrate can be reliably exposed to improve the connection reliability at the time of packaging the semiconductor device.

According to still another aspect of the invention, there is provided a method of manufacturing a resin-sealed type semiconductor device, comprising the steps of: preparing a lead frame in which a plurality of leads are arranged around a tub for supporting a semiconductor chip; bonding the tub and the semiconductor chip; connecting the surface electrodes of the semiconductor chip and the corresponding ones of the leads by connecting members; forming a sealing body by resin-sealing the semiconductor chip and the connecting members on the lead frame on the semiconductor chip arranging side, to arrange the plurality of leads on the face of the semiconductor device mounting side with the individual inner portions of the adjoining ones of the leads being separated; exposing the plurality of leads by polishing the face of the sealing body on the semiconductor device mounting side; and cutting and separating the individual leads from the frame portion of the lead frame by pressing, after the lead frame is placed on a cutting die with the face of the sealing body on the semiconductor device mounting side being directed upward, the leads under pressure with the cutting blade of the cutting die from the side of the mounted face of the leads.

According to a further aspect of the invention, there is provided a method of manufacturing a resin-sealed type semiconductor device, comprising the steps of: preparing a lead frame which is made of a copper alloy and in which a plurality of leads are arranged around a tub for supporting a semiconductor chip; bonding the tub and the semiconductor chip; connecting the surface electrodes of the semiconductor chip and the corresponding ones of the leads by bonding wires; forming a sealing body by resin-sealing the semiconductor chip and the wires on the lead frame on the semiconductor chip arranging side, to arrange the plurality of leads on the face of the semiconductor device mounting side with the individual inner portions of the adjoining ones of the leads being separated; exposing the plurality of leads by polishing the face of the sealing body on the semiconductor device mounting side with a brush which is made of a polyamide resin having abrasive grain thereon; forming a soldered layer on the mounted faces of the leads, as exposed from the face on the semiconductor device mounting side; and cutting and separating the individual leads from the frame portion of the lead frame by pressing, after the lead frame is placed on a cutting die with the face of the sealing body on the semiconductor device mounting side being directed upward, the leads under pressure with the cutting blade of the cutting die from the side of the mounted face of the leads.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a perspective view showing one embodiment of the structure of a semiconductor device (QFN) according to the invention;

FIG. 2 is a bottom view showing the structure of the QFN shown in FIG. 1 ;

FIG. 3 is a sectional view showing the structure of the QFN shown in FIG. 1 ;

Of FIGS. 4A to 4 C showing one example showing the package state in which the QFN shown in FIG. 1 is mounted on a mount substrate: FIG. 4A is a sectional view of a portion; FIG. 4B is an enlarged sectional view, as taken in a lead longitudinal direction, of a portion of a soldered portion; and FIG. 4C is an enlarged sectional view, as taken in a lead transverse direction, of a portion of the soldered portion;

FIG. 5 is a flow chart showing a manufacture method of one example of an assembling procedure in a method of manufacturing the QFN shown in FIG. 1 ;

FIGS. 6A , 6 B, 6 C, 6 D, 6 E and 6 F are sectional views showing one example of the structures of intermediates to be treated at the individual steps of the manufacture method shown in FIG. 5 ;

FIG. 7 is a sectional view showing one example showing the state in which a resin burr is formed on the back face of a sealing body after the molding step of the manufacture method of the QFN shown in FIG. 1 ;

FIG. 8 is a top plan view showing the structure of a matrix frame or one example of a lead frame to be used for manufacturing the QFN shown in FIG. 1 ;

FIG. 9 is a top plan view showing a portion of one example of the structure of a unit lead frame in the matrix frame shown in FIG. 8 ;

FIG. 10 is a sectional view showing a portion of one example of a wiring method in the wire bonding step of the manufacture of the QFN shown in FIG. 1 ;

FIG. 11 is a sectional view showing a portion of one example of a molded state in the manufacture of the QFN shown in FIG. 1 ;

Of FIGS. 12A to 12 C showing one example of the state of the matrix frame molded in the manufacture of the QFN shown in FIG. 1 ; FIG. 12A is an enlarged top plan view showing a portion; FIG. 12B is a sectional view taken along line A—A of FIG. 12A ; and FIG. 12C is a sectional view taken along line B—B of FIG. 12A ;

FIG. 13 is a sectional view showing a portion of one example of the polished sate in which a sealing body is polished on its back face in the manufacture of the QFN shown in FIG. 1 ;

FIG. 14 is a top plan view showing the example of the polished sate in which the sealing body is polished on its back face in the manufacture of the QFN shown in FIG. 1 ;

FIGS. 15A , 15 B, 15 C and 15 D are sectional views showing portions of one example of a lead cutting procedure in the manufacture of the QFN shown in FIG. 1 ;

FIG. 16 is a sectional view showing a flatness correcting method according to an application of the semiconductor device manufacturing method of the invention;

FIG. 17 is a sectional view showing a package thickness adjusting method according to an application of the semiconductor device manufacturing method of the invention;

FIG. 18 is a sectional view showing a structure of a semiconductor device (QFN) according to another embodiment of the invention; and

Of FIGS. 19A and 19B showing a molding method and a structure of a semiconductor device (QFN) to be compared with the semiconductor device of the invention, FIG. 19A is a sectional view showing a portion of a molding state, and FIG. 19B is a sectional view of the QFN to be compared.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

In the following embodiments, the description of identical or similar portions will not be repeated on principle, unless otherwise especially necessary.

Moreover, the following embodiments will be described, if needed for conveniences, by dividing them into a plurality of sections or modes, which are not independent of one another, unless otherwise explicitly specified so, but are partially or wholly in relations of modifications, details or supplementary explanations of others.

In the following embodiments, on the other hand, when reference is made to the number (including the numbers, numerical values, quantities and ranges) of elements, the invention should not be limited to the specific number, unless otherwise explicitly specified or apparently limited to that number, but may be more or less the specific number.

The embodiments of the invention will be described in detail with reference to the accompanying drawings. Here, throughout all the drawings for explaining the embodiments, parts having identical functions will be designated by common reference numerals, and their repeated descriptions will be omitted.

FIG. 1 is a perspective view showing one embodiment of the structure of a semiconductor device (QFN) according to the invention; FIG. 2 is a bottom view showing the structure of the QFN shown in FIG. 1 ; FIG. 3 is a sectional view showing the structure of the QFN shown in FIG. 1 ; Of FIGS. 4A to 4 C showing one example showing the package state in which the QFN shown in FIG. 1 is mounted on a mount substrate: FIG. 4A is a sectional view of a portion; FIG. 4B is an enlarged sectional view, as taken in a lead longitudinal direction, of a portion of a soldered portion; and FIG. 4C is an enlarged sectional view, as taken in a lead transverse direction, of a portion of the soldered portion; FIG. 5 is a flow chart showing a manufacture method of one example of an assembling procedure in a method of manufacturing the QFN shown in FIG. 1 ; FIGS. 6A , 6 B, 6 C, 6 D, 6 E and 6 F are sectional views showing one example of the structures of intermediates to be treated at the individual steps of the manufacture method shown in FIG. 5 ; FIG. 7 is a sectional view showing one example showing the state in which a resin burr is formed on the back face of a sealing body after the molding step of the manufacture method of the QFN shown in FIG. 1 ; FIG. 8 is a top plan view showing the structure of a matrix frame or one example of a lead frame to be used for manufacturing the QFN shown in FIG. 1 ; FIG. 9 is a top plan view showing a portion of one example of the structure of a unit lead frame in the matrix frame shown in FIG. 8 ; FIG. 10 is a sectional view showing a portion of one example of a wiring method in the wire bonding step of the manufacture of the QFN shown in FIG. 1 ; FIG. 11 is a sectional view showing a portion of one example of a molded state in the manufacture of the QFN shown in FIG. 1 ; Of FIGS. 12A to 12 C showing one example of the state of the matrix frame molded in the manufacture of the QFN shown in FIG. 1 ; FIG. 12A is an enlarged top plan view showing a portion; FIG. 12B is a sectional view taken along line A—A of FIG. 12A ; and FIG. 12C is a sectional view taken along line B—B of FIG. 12A ; FIG. 13 is a sectional view showing a portion of one example of the polished sate in which a sealing body is polished on its back face in the manufacture of the QFN shown in FIG. 1 ; FIG. 14 is a top plan view showing the example of the polished sate in which the sealing body is polished on its back face in the manufacture of the QFN shown in FIG. 1 ; and FIGS. 15A , 15 B, 15 C and 15 D are sectional views showing portions of one example of a lead cutting procedure in the manufacture of the QFN shown in FIG. 1 .

The semiconductor device of this embodiment, as shown in FIGS. 1 to 3 , is of a small resin-sealed type and a surface mount type. This embodiment will be described by taking up a QFN (Quad Flat Nonleaded Package) 7 as one example of the semiconductor device.

On the other hand, the QFN 7 is a small-sized semiconductor package, as incorporated mainly into a mobile electronic device, and is of a surface mount, in which a plurality of leads la connected with a land 9 a, i.e., substrate side terminals of a mount substrate 9 (as referred to FIG. 4 ) such as a printed-circuit board are arranged on the face of the semiconductor device mounting side (as will be called the “back face 3 a ”) of a sealing body 3 made of a mold, as shown in FIG. 2 .

Here, a semiconductor chip 2 , as shown in FIG. 3 , to be incorporated into the QFN 7 is a chip for a microcomputer or an ASIC (Application Specific Integrated Circuit), but should not be limited thereto.

Here will be described the construction of the QFN 7 (semiconductor device) of this embodiment with reference to FIGS. 1 to 4 .

The QFN 7 is constructed to include: a tub 1 e supporting the semiconductor chip 2 having a semiconductor integrated circuit formed on its principal face 2 b; the sealing body 3 formed by sealing the semiconductor chip 2 with a resin; the plurality of leads 1 a made of a copper alloy, arranged around the tub 1 e, exposed to the back face 3 a of the sealing body 3 , and having a soldered or plated layer 8 formed at the exposed portion (or a mounted face 1 d ); and bonding wires 4 (or connecting members) connecting pads 2 a or the surface electrodes of the semiconductor chip 2 with the corresponding leads 1 a. As shown in FIG. 4C , two widthwise edge portions 1 c, as exposed from the back face 3 a of the sealing body 3 , of the lead la are formed into rounded faces 1 g, and the central portion, as including the rounded faces 1 g, of the mounted face 1 d of the lead 1 a is protruded from the back face 3 a of the sealing body 3 .

Here, the protrusion of the lead 1 a from the back face 3 a of the sealing body 3 is achieved at a polishing step S 4 , as shown in FIG. 5 , of a manufacture method of the QFN 7 by polishing the back face 3 a of the sealing body 3 with a brush 10 which is made of a polyamide resin having abrasive gain adhered thereon, as shown in FIG. 13 .

When the back face 3 a of the sealing body 3 is polished with the brush 10 having abrasive grain, not only the resin of the back face 3 a of the sealing body 3 but also the longitudinal edge portions 1 c of the lead 1 a are polished, as shown in FIG. 4C , so that the edge portions 1 c become the rounded faces 1 g.

By comparing the thickness (T 1 ) of the central portion of the lead 1 a in the widthwise section of the lead 1 a with the thickness (T 2 ) of the end portions of the lead 1 a, therefore, the lead 1 a can be formed to satisfy T 1 >T 2 . After the polishing step, at a plating step S 5 shown in FIG. 5 , the mounted face 1 d of the lead 1 a is plated with solder so that the soldered layer 8 shown in FIG. 3 is formed on the mounted face 1 d.

Here, the protrusion (T 1 −T 2 ) at the central portion is about 1 to 3 microns.

In the QFN 7 of this embodiment, on the other hand, the tub 1 e has a thickness of about one half as large as that of the thickness of the lead 1 a, as shown in FIG. 3 .

Thus, at the time of manufacturing a matrix frame 1 (or lead frame) shown in FIG. 8 , the back face 1 h (as referred to FIG. 3 ) of the tub 1 e is etched off so that a mold resin 15 shown in FIG. 11 can be arranged on the side of the back face 1 h of the tub 1 e, too, without elevating the position of the tub 1 e with respect to the sealing body 3 .

Since the vertical position of the tub 1 e is unchanged in the QFN 7 shown in FIG. 3 , therefore, the QFN 7 can be thinned, and the tub 1 e is completely buried in the sealing body 3 so that the substrate can also be wired at positions on the lower side of the sealing body 3 in the mount substrate 9 when the QFN 7 is mounted on the mount substrate 9 , as shown in the packaged mode of FIG. 4 A. Here on the surface of the mount substrate 9 , there is formed a solder resist 9 b or an insulating film for covering the substrate wires.

In the QFN 7 , on the other hand, the semiconductor chip 2 is supported by the tub 1 e, which is exemplified in this embodiment by an X-shaped small cross tub, as shown in FIG. 9 . In other words, the tub 1 e is made far smaller in size than the semiconductor chip 2 .

As a result, the semiconductor chips 2 of a plurality of kinds of sizes can be packaged on the tub 1 e of one kind, and the contact area of the semiconductor chip 2 with the mold resin 15 on the side of its back side 2 c can be enlarged to improve the contact between the semiconductor chip 2 and the sealing body 3 thereby to prevent the sealing body 3 and the semiconductor chip 2 from being separated.

In the QFN 7 of this embodiment, on the other hand, a lead burr 1 i shown in FIG. 4B , as formed on the lead 1 a at the step S 6 of cutting the lead 1 a, as shown in FIG. 5 , is formed at the end portion on the upper side of the lead 1 a, i.e., on the face of the lead 1 a, as opposed to the mounted face 1 d.

When the lead 1 a is cut off the matrix frame 1 at the cutting step by the cutting punch 12 a (or cutting blade) of a cutting die 12 shown in FIG. 15 , it is pressed and cut with a shearing force by the cutting punch 12 a from the side of its mounted face 1 d of the lead 1 a. As a result, the lead burr 1 i is formed at the upper side end portion of the lead 1 a.

On the other hand, the semiconductor chip 2 is given a thickness of about 0.2 to 0.3 mm, for example, and is fixed on the tub 1 e with a die bond material 5 such as an epoxy adhesive.

On the other hand, the lead frame having the tub 1 e and the lead 1 a, i.e., the matrix frame 1 is made of a copper (Cu) alloy, an alloy (Fe—Ni) of iron and nickel or the like but is preferably made of the copper alloy.

Moreover, the matrix frame 1 has a thickness of about 0.09 to 0.21 mm, for example. Therefore, the tub 1 e or the lead 1 a has a similar thickness such as about one half as large as that of the lead 1 a when the tub 1 e is half-etched as in the QFN 7 of this embodiment.

Here, the matrix frame 1 should not be limited to the above-specified materials and thicknesses.

On the other hand, the bonding wire 4 (or the connecting member) for connecting the pad 2 a of the semiconductor chip 2 with the corresponding lead 1 a is made of a gold wire, for example.

Moreover, the sealing body 3 is formed by sealing it with a resin according to the molding method, and the used sealing mold resin 15 shown in FIG. 11 is a thermoset epoxy resin, for example.

Here, the sealing body 3 made of the mold resin 15 has a thickness of about 0.5 to 0.95 mm, for example.

On the other hand, the plated layer to be formed on the mounted face 1 d of each lead 1 a is the soldered layer 8 (Sn—Pb), for example, in this embodiment but may be another Sn solder such as a solder of Sn—Ag—Bi, a solder of Sn—Ag—Cu or a palladium (Pd) solder.

The plating thickness is 0.05 mm or less, for example, and is preferably about 0.01 mm.

Here, this plating thickness is the stand-off extent (or stand-off height) in the QFN 7 .

Where the thickness of the sealing body 3 is about 0.5 to 0.95 mm and where the thickness of the soldered layer 8 is about 0.01 mm, for example, the height of the mount substrate 9 from the land 9 a when the QFN 7 is packaged can be suppressed to 1.0 mm or less.

According to the QFN 7 of this embodiment, by polishing the back face 3 a of the sealing body 3 after the sealing body was formed by the resin mold, a resin burr 16 (of the leakage resin), as formed on the back face 3 a of the sealing body 3 and shown in FIG. 6C , and the resin burr 16 (or the resin flash), as shown in FIG. 7 , can be removed to expose the mounted face 1 d of each lead 1 a sufficiently. This makes it possible to expose the connection region of the lead 1 a, as needed when the QFN 7 is mounted on the mount substrate 9 , without fail.

In short, it is possible to retain the connection region of the lead 1 a, as needed at the time of packaging the QFN 7 .

As a result, the solder wettability can be improved to improve the connection reliability at the time when the QFN 7 is mounted on the mount substrate 9 or the like.

On the other hand, the flatness of the back face 3 a of the sealing body 3 can be improved by polishing it thereby to improve the packageability of the QFN 7 .

In the back face 3 a of the sealing body 3 , moreover, the rounded faces 1 g are formed at the longitudinal edge portions 1 c of the lead 1 a, as shown in FIG. 4C , so that the thickness (T 1 ) of the central portion of the lead 1 a in the widthwise section of the lead 1 a can be made larger than the thickness (T 2 ) of the end portions of the lead 1 a (T 1 >T 2 ).

As a result, the stand-off extent at the time of packaging the QFN 7 can be retained.

Since the widthwise central portion of the lead 1 a is thicker than the end portions, more specifically, a solder paste 14 can run sufficiently to the rounded faces 1 g at the widthwise end portions of the lead 1 a, as shown in FIG. 4C , when the mount substrate 9 is soldered.

Therefore, solder fillets in the solder connection can be sufficiently formed to improve the connection reliability at the time of mounting the QFN 7 on the mount substrate 9 .

In the QFN 7 of this embodiment, on the other hand, the lead burr 1 i is formed on the upper side of the lead 1 a, as shown in FIG. 4 B.

As a result, the mounted face 1 d of the lead 1 a of the QFN 7 does not have lead burr 1 i formed but is flat so that the packageability at the time when the QFN 7 is mounted on the mount substrate 9 and the connection reliability at the soldering time can be improved.

Since the matrix frame 1 is made of the copper alloy, on the other hand, its material cost can be lowered, and the heat radiation of the QFN 7 can be improved.

Therefore, it is possible to improve the reliability of the QFN 7 .

On the other hand, the soldered layer 8 is formed as a plated layer on the mounted face 1 d of the lead 1 a of the QFN 7 so that the solder connectability at the time of mounting the QFN 7 on the mount substrate 9 can be improved.

Where the palladium plated layer is formed as the plated layer, the QFN 7 can be set free of lead when it is manufactured.

Next, the method of manufacturing the QFN 7 of this embodiment will be described with reference to the flow chart of the manufacture method shown in FIG. 5 .

First of all, there is prepared the matrix frame 1 (or the lead frame), as shown in FIG. 8 , which is made of the copper alloy with the plurality of leads 1 a being arranged around the tub 1 e capable of supporting the semiconductor chip 2 .

Here in this matrix frame 1 , unit lead frames 1 j for the individual package regions are formed in a plurality of rows and a plurality of columns so that a number of QFNs 7 can be manufactured from one matrix frame 1 . FIG. 9 shows the structure of the unit lead frame 1 j of the matrix frame 1 of FIG. 8 in detail.

In the matrix frame 1 of this embodiment, on the other hand, of the plurality of leads 1 a, the individual inner portions 1 b of the adjoining leads 1 a are separated in advance. After the resin molding step, the lead root end portions, as slightly protruding from the sealing body 3 , are connected and supported by a frame portion 1 f.

Here in the matrix frame 1 , as shown in FIG. 8 , a plurality of guide holes 1 k are formed on the two side portions for providing guides at the transferring time and at the positioning time.

As shown in FIG. 9 , moreover, at each unit lead frame 1 j, the tub 1 e is supported in four directions by tub suspending leads 11 , and slits 1 m are formed around the individual leads 1 a for absorbing the stress.

After this, there is prepared the semiconductor chip 2 having the semiconductor integrated circuit formed on the principal face 2 b. The semiconductor chip 2 is subjected to the die bonding (or chip mounting or pellet bonding) treatment at step S 1 .

Specifically, the tub 1 e in each unit lead frame 1 j of the matrix frame 1 and the semiconductor chip 2 are bonded.

At this time, the die bonding material 5 such as silver paste is applied to the tub 1 e, as shown in FIG. 6A , to bond the tub 1 e and the back face 2 c of the semiconductor chip 2 through that die bonding material 5 .

In other words, the semiconductor chip 2 is fixed over the tub 1 e through the die bonding material 5 with its principal face 2 b directed upward.

After this, there is performed a wire bonding treatment, as indicated at step S 2 .

First of all, the matrix frame 1 is set on a heat stage 17 , as shown in FIG. 10 , and the pads 2 a of the semiconductor chip 2 and the inner portions 1 b of the corresponding leads 1 a are connected by the wires 4 of the gold line, as shown in FIG. 6B , by using a capillary 6 or a bonding tool.

At this time, in this embodiment, the pads 2 a of the semiconductor chip 2 and the wires 4 are connected at first on the first bond side, and the wires 4 and the inner portions 1 b of the leads 1 a are then connected on the second bond side.

After this, a molding treatment is performed, as indicated at step S 3 . Here, the molding treatment is done by the transfer molding method.

At first, the wire-bonded matrix frame 1 (as referred to FIG. 10 ) is set on the upper part 11 a of a mold die 11 , as shown in FIG. 11 , and is clamped by the upper part 11 a and a lower part 11 b. After this, the mold resin 15 is injected into a cavity 11 c which is defined by the upper part 11 a and the lower part 11 b.

Here in the matrix frame 1 used in this embodiment, the individual inner portions 1 b of the adjoining ones of the plurality of leads 1 a are separated in advance. By resin-molding the semiconductor chip 2 and the wires 4 on the arranged side (although the back face 1 h of the tub 1 e is included) of the semiconductor chip 2 on the matrix frame 1 , therefore, the plurality of leads 1 a are arranged on the back face 3 a to form the sealing body 3 with the individual inner portions 1 b of the adjoining leads 1 a being separate.

Therefore, the mold of this embodiment is the one-side mold which uses the upper part 11 a having no recessed cavity 11 c formed.

The mold resin 15 is set, when charged up, after lapse of a predetermined time period, and the mold is then opened to extract the matrix frame 1 from the mold die 11 .

Here, FIGS. 12A , 12 B and 12 C show the states of the matrix frame 1 after molded. In the matrix frame 1 extracted from the mold die 11 , as shown in FIGS. 12A and 12C , a residual resin such as a gate 18 or a runner 19 is formed together with the plurality of sealing bodys 3 .

Here in the molded matrix frame 1 , on the back face 3 a of each sealing body 3 , there are formed the resin burr 16 of the leakage resin, as shown in FIG. 6C , and the resin burr 16 of the resin flash, as shown in FIG. 7 . At this stage, the connection region for the mounted faces 1 d of the leads 1 a is not sufficiently retained on the back face 3 a of the sealing body 3 .

After this, the polishing treatment is performed, as indicated at step S 4 . Specifically, the back face 3 a of each sealing body 3 in the matrix frame 1 is polished to expose the mounted face 1 d of each lead 1 a sufficiently for retaining the connection region of the mounted face 1 d sufficiently.

At this time, in this embodiment, the back side 3 a of the sealing body 3 is polished with the brush 10 , as made of the polyamide resin having the abrasive grain and as shown in FIG. 13 , to expose the plurality of leads 1 a to the outside.

Here, the brush 10 is made of the resin having the abrasive grain thereon and is preferably made of a polyamide resin. However, the brush 10 may be made of a nonwoven fabric and is borne turnably and slidably.

For this polishing treatment, the molded matrix frame 1 is set on a mask jig 21 which is attached to a polishing stage 20 .

Here, the mask jig 21 is provided in a matrix arrangement with a plurality of through holes 21 a according to the positions and sizes of the plurality of sealing bodys 3 of the matrix frame 1 .

When the matrix frame 1 is set on the polishing stage 20 , therefore, its face on the side having the sealing bodys 3 is directed downward, as shown in FIG. 13 , so that the individual sealing bodys 3 are arranged in the individual through holes 21 a of the mask jig 21 placed on the polishing stage 20 .

As a result, the sealing bodys 3 have their back faces 3 a directed upward on the polishing stage 20 .

After this, an evacuation is made from the side of the back face 21 b of the mask jig 21 to fix the matrix frame 1 on the mask jig 21 .

Subsequently, the polishing treatment is performed by turning the brush 10 at a predetermined speed while discharging cooling water 23 from nozzles 22 to the back faces 3 a of the individual sealing bodys 3 and simultaneously by moving the brush 10 in a predetermined direction.

Here in the moving locus of the brush 10 , for example, the brush 10 is preferably moved so uniformly over the matrix frame 1 as to come into substantially homogeneous contact with the individual sealing bodys 3 of the matrix frame 1 . In this embodiment, as shown in FIG. 14 , the brush 10 is reciprocated, as indicated at 24 , in parallel with the longitudinal direction of the matrix frame 1 . However, the moving locus of the brush 10 should not be limited to that shown in FIG. 14 but may be any if it is homogeneous over the matrix frame 1 .

During the movement, moreover, the brush 10 continuously applies a predetermined load to the back faces 3 a of the individual sealing bodys 3 thereby to polish the back faces 3 a.

Here, the abrasion of the back faces 3 a of the sealing bodys 3 is about 1 to 3 microns, for example.

As a result, the resin burr 16 , as shown in FIG. 6C or 7 , can be removed to expose the mounted face 1 d of each lead 1 a so sufficiently on the back face 3 a of the sealing body 3 as to retain the connection region as shown in FIG. 6 D.

By polishing with the brush 10 made of the polyamide resin having the abrasive grain, moreover, the mounted face 1 d of the lead 1 a can be so shaped that the longitudinal edge portions 1 c of the lead 1 a may be the rounded faces 1 g, as shown in FIG. 4 C. As a result, the thickness (T 1 ) of the central portion of the lead 1 a in the widthwise section of the lead 1 a can be made larger than that (T 2 ) of the end portions of the lead 1 a (T 1 >T 2 ).

After this, there is performed the plating treatment, as indicated at step S 5 .

Specifically, the mounted face 1 d of each lead 1 a, as exposed from the back face 3 a of the sealing body 3 , is plated with the soldered layer 8 (or the plated layer), as shown in FIG. 6 E.

At this time, the soldered layer 8 is formed by the electrolytic method, for example, to have a thickness of about 0.05 mm (at the maximum).

After this, there is performed the cutting treatment, as indicated at step S 6 .

At the cutting step, as shown in FIG. 15A , the matrix frame 1 is placed on a lead cutting die 12 b of the cutting die 12 with the back face 3 a of the sealing body 3 being directed upward. Subsequently, the matrix frame 1 is clamped by the lead cutting die 12 b and a stripper 12 c.

After this, as shown in FIG. 15B , the lead 1 a is pressed (or given a shearing force) downward or from the side of the mounted face 1 d by the cutting punch 12 a (or cutting blade) of the cutting die 12 so that it is cut off the frame portion 1 f, as shown in FIG. 9 , of the matrix frame 1 so that the individual leads 1 a are separated from the frame portion 1 f.

Here, by pressing the cutting punch 12 a downward of the lead 1 a, foreign substances such as cut chips 13 are enabled to fall down, as shown in FIG. 15C , and the lead burr 1 i can be formed on the face of the lead 1 a on the side opposed to the mounted face 1 d, as shown in FIG. 15 D and FIG. 4 B.

Thus, there can be assembled the QFN 7 , as shown in FIGS. 1 to 3 and FIG. 6 F.

On the other hand, the packaged state of the completed QFN 7 on the mount substrate 9 is shown in FIG. 4 .

According to the manufacture method of the QFN 7 (or the semiconductor device) of this embodiment, the resin burr 16 , as formed on the back face 3 a of the sealing body 3 , can be removed by polishing the back face 3 a of the sealing body 3 after the resin molding step so that the mounted face 1 d of the lead 1 a can be reliably exposed.

As a result, the connection region (or the plated region) on the mounted face 1 d of the lead 1 a at the time of packaging the QFN 7 can be retained to improve the connection reliability of the QFN 7 .

Since the mounted face 1 d of the lead 1 a can be reliably exposed to the back face 3 a by polishing the back face 3 a of the sealing body 3 after the resin molding step, on the other hand, it is possible to manufacture even the QFN 7 of the large size (having 100 pins or more), although having been relatively difficult to manufacture.

By polishing the back face 3 a of the sealing body 3 after the resin molding step, moreover, the resin burr 16 can be removed from the back face 3 a. This makes it unnecessary to manage the pressure of the mold die 11 , as might otherwise be caused by the resin burr 16 such as the resin flash, so that the method management in the molding step can be facilitated.

Since the resin burr 16 is removed from the back face 3 a of the sealing body 3 , on the other hand, the frequency to clean the mold die 11 can be reduced to improve the throughput of the molding step.

After the resin molding step, on the other hand, the mounted face 1 d of the lead 1 a can be reliably exposed to the back face 3 a of the sealing body 3 by polishing the back face 3 a so that the laminate molding method is not needed. This makes it unnecessary to modify the molding apparatus so that no cost is included for the modification.

Without the laminate molding method, moreover, the clamp is not made defective, as might otherwise occur at the lead clamping time for cutting the lead, so that the lead can be prevented from becoming defective.

Since the back face 3 a of the sealing body 3 is polished by using the brush 10 made of the polyamide resin, on the other hand, the polishing treatment can be performed at an abrasion rate of about several microns according to the warpage (or deformation), if any, of the back face 3 a of the sealing body 3 .

As a result, the lead 1 a can be exposed by polishing the back face 3 a of the sealing body 3 while keeping the shape of its back face 3 a.

Since the back face 3 a of the sealing body 3 is polished by using the brush 10 made of the polyamide resin, on the other hand, the longitudinal edge portions 1 c of the lead 1 a on the back face 3 a of the sealing body 3 can be formed into the rounded faces 1 g, and the thickness (T 1 ) of the central portion of the lead 1 a can be made larger than that (T 2 ) of the end portions of the lead 1 a (T 1 >T 2 ).

As a result, the widthwise thickness of the central portion of the lead 1 a is made larger than that of the end portions so that the solder paste 14 can run sufficiently to the rounded faces 1 g at the widthwise end portions of the lead 1 a, as shown in FIG. 4C , when the mount substrate 9 is soldered. As a result, the solder fillet at the soldered connection can be sufficiently formed to enhance the connection reliability at the time when the QFN 7 is mounted on the mount substrate 9 or the like.

At the cutting step of the lead 1 a, on the other hand, the foreign substances such as the cut chips 13 can be dropped down by bringing the cutting punch 12 a downward (from the side of the mounted face 1 d ) under the pressure into contact with the lead 1 a.

As a result, the foreign substances such as the cut chips 13 can be efficiently collected to improve the working efficiency of the cutting step.

By bringing the cutting punch 12 a downward (from the side of the mounted face 1 d ) under the pressure into contact with the lead 1 a, on the other hand, the foreign substances go around to the face opposed to the mounted face 1 d of the lead 1 a so that the mounted face 1 d can be set free of the foreign substances.

Thus, it is possible to improve the connection reliability at the time of packaging the QFN 7 .

By bringing the cutting punch 12 a downward (from the side of the mounted face 1 d ) under the pressure into contact with the lead 1 a, moreover, the lead burr 1 i can be formed on the faces, as opposed to the mounted face 1 d, of the lead 1 a.

As a result, the mounted face 1 d of the lead 1 a can be flatted without any burr to improve the connection reliably at the time of packaging the QFN 7 like before.

Although our invention has been specifically described in connection with its embodiments, it should not be limited to those embodiments but could naturally be modified in various manners without departing from the gist thereof.

In connection with the embodiments, for example, there has been described the technique for removing the resin burr 16 such as the leakage resin or the resin flash, as formed on the back face 3 a of the resin-molded sealing body 3 , by polishing the back face 3 a. An effect other than that for removing the resin burr 16 (as referred to FIG. 6C ) could be obtained, as exemplified in other embodiments shown in FIGS. 16 and 17 .

In the embodiment of FIG. 16 , more specifically, the flatness of the back face 3 a of the resin-molded sealing body 3 is adjusted by polishing the back face 3 a. As a result, the back face 3 a can be cleared of the resin burr 16 and can be easily given a desired flatness.

In FIG. 17 , moreover, the resin-molded sealing body 3 is thinned by polishing its back face 3 a. As a result, the back face 3 a can be cleared of the resin burr 16 , and the QFN 7 (as referred to FIG. 1 ) can be made thinner.

By the semiconductor device manufacturing method according to the foregoing and other embodiments, therefore, it is possible to deburr the back face 3 a of the QFN 7 , to flatten the back face 3 a and to thin the QFN 7 .

On the other hand, the foregoing embodiment has been described on the case in which the QFN 7 has the tub-buried structure. In another embodiment shown in FIG. 18 , however, the QFN 7 may be modified into the tub-exposed structure in which the tub 1 e is exposed to the back side 3 a of the sealing body 3 .

In this case, the back face 3 a of the sealing body 3 is polished after molded with the resin, to expose the individual leads 1 a and the tub 1 e to the back face 3 a of the sealing body 3 . In the QFN 7 having the tub-exposed structure shown in FIG. 18 , too, there can be attained an effect similar to that of the QFN 7 of the tub-buried structure, as has been described in connection with the foregoing embodiment.

Moreover, the foregoing embodiment has been described on the case in which the tub 1 e is the X-shaped small cross tub. However, the shape of the tub 1 e should not be especially limited but may be rectangular or circular.

Moreover, the tub 1 e should not be limited to the small tub but may be of the type larger than the semiconductor chip 2 .

On the other hand, the foregoing embodiment has been described on the case in which the lead frame is the matrix frame 1 . However, the lead frame may be a series of unit lead frames 1 j arranged in a row.

Moreover, the foregoing embodiment has been described on the case in which the back face 3 a of the sealing body 3 is polished with the brush 10 . However, the method of deburring the back face 3 a of the QFN 7 , flattening the back face 3 a and thinning the QFN 7 may employ a grinding stop other than the brush 10 .

On the other hand, the foregoing embodiment has been described on the case in which the lead 1 a is press-cut by bringing the cutting blade (or the cutting punch 12 a ) under pressure from the side of the mounted face 1 d of the lead 1 a into contact. However, the cutting need not be performed from the side of the mounted face 1 d but may be effected from the opposite side.

In the method of manufacturing the semiconductor device of the invention, moreover, the technique of polishing the back side of the sealing body 3 and the technique of cutting the lead 1 a, as have been described in connection with the foregoing embodiment, may be performed both simultaneously or either solely.

Here, the foregoing embodiment has been described on the case in which the semiconductor device is the small-sized QFN 7 . However, the semiconductor device may be any other than the QFN 7 if it is of the resin-sealed type and of the surface mount type in which it is assembled by using the lead frame.

The effects to be obtained by the representative invention disclosed herein will be briefly described in the following.

(1) By polishing the back face of the resin-molded sealing body, it is possible to remove the resin burr which is formed on the back face of the sealing body. As a result, the mounted face of the lead can be reliably exposed. Therefore, the connection region (or the plated region) can be retained on the mounted face of the lead at the time of packaging the semiconductor device. As a result, it is possible to improve the connection reliability of the semiconductor device.

(2) By polishing the back face of the resin-molded sealing body, it is possible to remove the resin burr which is formed on the back face of the sealing body. Therefore, the pressure management of the mold die, as might otherwise be caused by the resin burr such as the resin flash burr, can be eliminated to make it easy to manage the molding step.

(3) Since the mounted face of the lead can be reliably exposed by polishing the back face of the resin-molded sealing body, no laminate molding method is needed to require no modification of the molding apparatus so that no cost is involved for the modification. No laminate molding method is performed to cause no defect at the time of clamping the lead when the lead is cut, so that the lead cutting defect can be prevented.

(4) By polishing the back face of the sealing body with the brush made of the polyamide resin, the back face of the sealing body can be polished, even if deformed or warped, at the abrasion rate of about several microns according to the warpage of that back face.

(5) By polishing the back face of the sealing body with the brush made of the polyamide resin, the longitudinal edge portions of the lead on the back face of the sealing body can be formed into the rounded faces. Therefore, the widthwise central portion of the lead can be made thicker than the end portions so that the solder paste can run sufficiently to the rounded faces of the widthwise end portions of the lead when the mount substrate or the like is soldered. Therefore, the solder fillet in the solder connection can be sufficiently formed to enhance the connection reliability at the time of packaging the semiconductor device.

(6) At the cutting step of the lead, the foreign substances such as the cut chips can be dropped down by bringing the cutting blade downward under the pressure into contact with the lead. As a result, the foreign substances can be efficiently collected to improve the working efficiency of the cutting step.

(7) By bringing the cutting blade downward under the pressure into contact with the lead, the foreign substances go around to the face opposed to the mounted face of the lead so that the mounted face can be set free of the foreign substances. Therefore, it is possible to improve the connection reliability at the time of packaging the semiconductor device. Moreover, the lead burr can be formed on the upper side of the lead thereby to improve the connection reliability at the time of packaging the semiconductor device.

Claims

1. A method of manufacturing a semiconductor device, comprising the steps of:providing a lead frame having a front face, a back face opposed of the front face, a tub and a plurality of leads arranged around the tub wherein a back face of the tub is etched off and each of the leads is thicker than the tub; providing a semiconductor chip having a principal face, back face and a plurality of electrodes formed on the principal face; bonding the back face of the semiconductor chip on the front face of the tub; electrically connecting the plurality of electrodes with the plurality of leads via a plurality of wires, respectively; providing a molding die having an upper part and a lower part wherein the lower part has a cavity; arranging the lead frame between the upper part and the lower part as the bonded semiconductor chip is positioned in the cavity of the lower part and the back face of the tub is facing the upper part; and after the arranging step, injecting a resin in the cavity for sealing the semiconductor chip, the tub, the plurality of wires, and a part of each of the plurality of leads in the resin, wherein the back face of the tub is entirely sealed by the resin.

2. A method of manufacturing a semiconductor device according to claim 1, wherein back faces of each of the plurality of leads are exposed from the resin.

3. A method of manufacturing a semiconductor device according to claim 2, wherein the front face of the tub is smaller than the back face of the semiconductor chip and a part of the back face of the semiconductor chip is in contact with the resin.

4. A method of manufacturing a semiconductor device according to claim 3, wherein a position of the tub of the provided lead frame is not an elevated position.