JP4443190B2 - Manufacturing method of semiconductor device - Google Patents

Description

  The present invention relates to a method for manufacturing a semiconductor device, and more particularly to a method for manufacturing a thin semiconductor device.

  2. Description of the Related Art Conventionally, a semiconductor device set in an electronic device is used in a mobile phone, a portable computer, and the like, and thus, a reduction in size, thickness, and weight are required.

  For example, as a general semiconductor device, there is a conventional package type semiconductor device sealed with a normal transfer mold. The semiconductor device 1 is mounted on a printed circuit board PS as shown in FIG.

  In the package type semiconductor device 1, the periphery of the semiconductor chip 2 is covered with a resin layer 3, and lead terminals 4 for external connection are led out from the sides of the resin layer 3.

  However, the package type semiconductor device 1 has lead terminals 4 protruding from the resin layer 3 and has a large overall size, which does not satisfy the miniaturization, thickness reduction, and weight reduction.

  Therefore, various companies have competed to develop various structures to achieve miniaturization, thinning, and weight reduction, and recently called CSP (chip size package), wafer scale CSP equivalent to chip size, or chip size A slightly larger CSP has been developed.

  FIG. 25 shows a CSP 6 that employs a glass epoxy substrate 5 as a support substrate and is slightly larger than the chip size. Here, description will be made assuming that the transistor chip T is mounted on the glass epoxy substrate 5.

  A first electrode 7, a second electrode 8 and a die pad 9 are formed on the surface of the glass epoxy substrate 5, and a first back electrode 10 and a second back electrode 11 are formed on the back surface. The first electrode 7 and the first back electrode 10 are electrically connected to the second electrode 8 and the second back electrode 11 through the through hole TH. Further, the bare transistor chip T is fixed to the die pad 9, the emitter electrode of the transistor and the first electrode 7 are connected via the fine metal wire 12, and the base electrode of the transistor and the second electrode 8 are connected to the fine metal wire 12. Connected through. Further, a resin layer 13 is provided on the glass epoxy substrate 5 so as to cover the transistor chip T.

  The CSP 6 employs the glass epoxy substrate 5, but unlike the wafer scale CSP, the extending structure from the chip T to the backside electrodes 10 and 11 for external connection is simple, and has an advantage that it can be manufactured at low cost.

  The CSP 6 is mounted on the printed circuit board PS as shown in FIG. The printed circuit board PS is provided with electrodes and wirings constituting an electric circuit, and the CSP 6, the package type semiconductor device 1, the chip resistor CR, the chip capacitor CC, and the like are electrically connected and fixed.

  And the circuit comprised with this printed circuit board is attached in various sets.

  Next, a method for manufacturing the CSP will be described with reference to FIGS. In FIG. 27, reference is made to a flow diagram entitled the central glass epoxy / flexible substrate.

First, a glass epoxy substrate 5 is prepared as a base material (support substrate), and Cu foils 20 and 21 are pressure-bonded to both surfaces via an insulating adhesive. (See FIG. 26A above)
Subsequently, the Cu foils 20 and 21 corresponding to the first electrode 7, the second electrode 8, the die pad 9, the first back electrode 10, and the second back electrode 11 are coated with an etching resistant resist 22, and Cu The foils 20 and 21 are patterned. Patterning may be performed separately for the front and back sides (see FIG. 26B above).
Subsequently, a hole for the through hole TH is formed in the glass epoxy substrate by using a drill or a laser, and the hole is plated to form the through hole TH. The first electrode 7 and the first back electrode 10, and the second electrode 8 and the second back electrode 10 are electrically connected through the through hole TH. (See FIG. 26C above)
Further, although omitted in the drawing, the first electrode 7 and the second electrode 8 that become bonding posts are plated with Ni, and the die pad 9 that becomes a die bonding post is plated with Au, and the transistor chip T is die bonded. To do.

Finally, the emitter electrode of the transistor chip T and the first electrode 7, the base electrode of the transistor chip T and the second electrode 8 are connected via the metal thin wire 12 and covered with the resin layer 13. (See FIG. 26D above)
If necessary, it is diced and separated as individual electric elements. In FIG. 26, only one transistor chip T is provided on the glass epoxy substrate 5, but actually, a large number of transistor chips T are provided in a matrix. Therefore, it is separated individually by a dicing device at the end.

  With the above manufacturing method, a CSP type electric element employing the support substrate 5 is completed. This manufacturing method is the same even if a flexible sheet is adopted as the support substrate.

On the other hand, a manufacturing method employing a ceramic substrate is shown in the flow on the left side of FIG. After preparing a ceramic substrate (green sheet) which is a support substrate, through holes are formed, and then the front and back electrodes are printed and sintered using a conductive paste. After that, until the resin layer of the previous manufacturing method is coated, the manufacturing method is the same as that shown in FIG. 26. However, the ceramic substrate is very brittle, and unlike a flexible sheet or glass epoxy substrate, it is chipped immediately. There is a problem that can not be molded. Therefore, the potting resin is potted and cured, and then polishing for flattening the sealing resin is performed, and finally, the dicing apparatus is used for individual separation.
JP 2001-339151 A (first page, FIG. 1)

  In FIG. 25, the transistor chip T, the connecting means 7 to 12 and the resin layer 13 are necessary components for electrical connection with the outside and protection of the transistor. It has been difficult to provide an electric circuit element that can be made thinner, thinner, and lighter.

  Moreover, the glass epoxy board | substrate 5 used as a support substrate is an essentially unnecessary thing as mentioned above. However, since the electrodes are bonded together in the manufacturing method, it is adopted as a support substrate, and the glass epoxy substrate 5 cannot be eliminated.

  For this reason, the use of the glass epoxy substrate 5 increases the cost. Further, since the glass epoxy substrate 5 is thick, it becomes thick as a semiconductor device, and there is a limit to miniaturization, thickness reduction, and weight reduction.

  Furthermore, the top of the thin metal wire 12 serving as the connection means is higher than the transistor chip T, and the semiconductor device 6 is accordingly thickened.

  In addition, a glass-epoxy substrate or a ceramic substrate always requires a through-hole forming process for connecting electrodes on both sides, and there is a problem that the manufacturing process becomes long.

  FIG. 28 shows a pattern diagram formed on a glass epoxy substrate, a ceramic substrate, a metal substrate, or the like. In this pattern, an IC circuit is generally formed, and a transistor chip 21, an IC chip 22, a chip capacitor 23, and / or a chip resistor 24 are mounted. A bonding pad 26 integrated with the wiring 25 is formed around the transistor chip 21 and the IC chip 22, and the chips 21 and 22 and the bonding pad 26 are electrically connected through a thin metal wire 28. The wiring 29 is formed integrally with the external lead pad 30. These wirings 25 and 29 extend while bending in the substrate, and are formed to be the thinnest in the IC chip if necessary. Therefore, these thin wirings have a problem that the bonding area with the substrate is very narrow and the wirings are peeled off or warped. The bonding pad 26 includes a power bonding pad and a small signal bonding pad. In particular, the small signal bonding pad has a small bonding area and causes film peeling.

  Furthermore, although the external lead is fixed to the external lead pad 30, there is a problem that the external lead pad 30 is peeled off by an external force applied to the external lead.

  The present invention is made in view of the above-described many problems, and includes a plurality of conductive paths electrically separated by a separation groove, a thin metal wire connecting the conductive path and the electrode of the semiconductor element, and the semiconductor element. The problem is solved by providing an insulating resin that covers and fills the separation groove between the conductive paths and supports the back surface of the conductive path and the back surface of the semiconductor element so as to be integrally supported.

  Also, a plurality of conductive paths electrically separated by the separation grooves, a metal thin wire connecting the conductive paths and the electrodes of the semiconductor elements, and the separation grooves between the conductive paths are filled with the semiconductor elements. The problem is solved by providing an insulating resin that exposes and integrally supports the conductive film formed on the back surface of the conductive path and the back surface of the semiconductor element.

Furthermore, a plurality of conductive paths electrically separated by the separation groove, a semiconductor element fixed on the first conductive path, a metal thin wire connecting the electrode of the semiconductor element and the second conductive path, An insulating resin that covers the semiconductor element and that is filled in the separation groove between the conductive paths and that integrally supports the exposed back surface of the conductive path;
The problem is solved by forming the first conductive path lower than the second conductive path.

  With this structure, the number of components can be minimized, and the top of the fine metal wire can be set low, so that the conventional problems can be solved.

Also, a conductive foil is prepared, and a groove shallower than the thickness of the conductive foil is formed in the conductive foil corresponding to the region where the semiconductor element is arranged between the conductive paths to be formed,
The semiconductor element is fixed to the groove in which the semiconductor element is disposed, and the electrode of the semiconductor element and the desired conductive path are electrically connected by a thin metal wire,
Cover the semiconductor element and the fine metal wire, mold with an insulating resin so as to fill the groove,
The problem is solved by separating the conductive paths to be formed.

Also, a conductive foil is prepared, and a groove shallower than the thickness of the conductive foil is formed by etching in the conductive foil corresponding to a region where the semiconductor element is disposed between the conductive paths to be formed, and the formation is performed by the etching. Form eaves with a material different from the conductive foil in the planned conductive path,
The semiconductor element is fixed in a groove in which the semiconductor element is disposed, and the electrode of the semiconductor element and the desired conductive path are electrically connected by a thin metal wire,
Cover the semiconductor element and the fine metal wire, mold with an insulating resin so as to fill the groove,
The problem is solved by separating the conductive paths to be formed.

  Through this manufacturing method, a through hole can be made unnecessary, and at the same time, the conductive foil is utilized as a supporting substrate and a conductive path, the number of components is minimized, and the conductive path does not escape from the insulating resin. . In addition, since the top of the thin metal wire can be set low by forming the groove, the thickness of the semiconductor device can be reduced.

  Furthermore, eaves are formed on the surface of the conductive path, and the conductive path is prevented from coming off by an insulating resin covering the eaves and filling the separation grooves.

  As is apparent from the above description, the present invention provides a semiconductor device that is configured with the minimum necessary amount of semiconductor elements, conductive paths, and insulating resin, and that does not waste resources. Therefore, a semiconductor device can be realized in which there are no extra components until completion and the cost can be significantly reduced. In addition, by setting the coating thickness of the insulating resin and the thickness of the conductive foil to the optimum values, and further setting the back surface of the semiconductor element to be lower than the surface of other conductive paths, the semiconductor device can be made thin. Therefore, it is possible to realize a semiconductor device that is reduced in size and weight. Furthermore, since the wiring in which the phenomenon of warping or peeling is remarkable is embedded and supported in an insulating resin, these problems can be solved.

  In addition, since only the back surface of the conductive path is exposed from the insulating resin, the back surface of the conductive path can be immediately used for connection to the outside, and the advantage that the back electrode and the through hole having the conventional structure as shown in FIG. 26 can be eliminated. Have

  In addition, since the semiconductor element is directly fixed to the conductive path through a conductive film such as brazing material, Au, Ag, or the like, or the back surface of the semiconductor element is exposed, the heat generated from the semiconductor element is directly transmitted through the conductive path. Heat can be transferred to the mounting board. In particular, the power element can be mounted by this heat radiation.

  Further, since the eaves can be formed on the surface of the conductive path, an anchor effect can be generated, and warping and disconnection of the conductive path, particularly the wiring can be prevented.

  In the method for manufacturing a semiconductor device of the present invention, the conductive foil itself functions as a support substrate, and the whole is supported by the conductive foil until the separation groove is formed or the semiconductor element is mounted and the insulating resin is applied. When separating the conductive foil as each conductive path, the insulating resin is used as a support substrate to function. Accordingly, the semiconductor element, conductive foil, and insulating resin can be manufactured with the minimum necessary. As described in the conventional example, a support substrate is not necessary when the semiconductor device is originally configured, and the cost can be reduced. In addition, a support substrate is not required, the conductive path is embedded in an insulating resin, and the thickness of the insulating resin and conductive foil can be adjusted, so that an extremely thin semiconductor device can be formed. There is also.

  In addition, since the through hole forming process, the conductor printing process (in the case of a ceramic substrate), and the like can be omitted, the manufacturing process can be greatly shortened compared to the prior art, and the entire process can be produced internally. Also, a frame mold is not required at all, and this is a manufacturing method with extremely short delivery time.

  Next, since the conductive paths can be handled without being individually separated, there is a feature that workability is improved in the subsequent coating step of the insulating resin.

  Finally, the semiconductor device can be used as a support substrate, and a semiconductor element can be mounted on the exposed conductive path, so that a highly functional substrate module can be realized. In particular, if the semiconductor device is used as a support substrate and the semiconductor device is mounted thereon as an element, a lighter and thinner substrate module can be realized.

First Embodiment Explaining Semiconductor Device First, the structure of a semiconductor device of the present invention will be described with reference to FIG.

  FIG. 1 includes a first conductive path 51A supported by an insulating resin 50, a second conductive path 51B and a third conductive path 51C embedded in the insulating resin 50, and the first conductive path 51C A semiconductor device 53 in which a semiconductor element 52 is fixed on the conductive path 51A is shown.

  This structure is composed of three materials: a semiconductor element 52A, a circuit element 52B, a plurality of conductive paths 51A, 51B, and 51C, and an insulating resin 50 that supports and embeds the conductive paths 51A, 51B, and 51C. A separation groove 54 filled with the insulating resin 50 is provided between the conductive paths 51.

  As the insulating resin, a thermosetting resin such as an epoxy resin, a thermoplastic resin such as a polyimide resin or polyphenylene sulfide can be used. As the insulating resin, any resin can be adopted as long as it is a resin that can be hardened using a mold, a resin that can be coated by dipping or coating. As the conductive path 51, a conductive foil mainly made of Cu, a conductive foil mainly made of Al, or a conductive foil made of an alloy such as Fe-Ni can be used. Of course, other conductive materials are possible, and a conductive material that can be etched and a conductive material that evaporates with a laser are particularly preferable.

  The connection means of the semiconductor element 52A is a thin metal wire 55A, a brazing material such as solder, or a conductive coating 55C such as an Ag paste or a conductive material. Further, the solder 55B is selected as the circuit element 52B such as a chip resistor or a chip capacitor.

  Further, for the adhesion between the semiconductor element 52A and the first conductive path 51A, an insulating adhesive is selected if an electrical connection is not required, and a conductive material 55C is employed if an electrical connection is required. . Here, at least one conductive material is sufficient.

  The material considered as the conductive material 55C is Ag, Au, Pt, Pd, or the like, which is coated by low vacuum or high vacuum deposition such as vapor deposition, sputtering, or CVD, plating, or sintering of a conductive paste. The

  For example, Ag adheres to Au and also to a brazing material. Therefore, if the Au film is coated on the back surface of the chip, the chip can be thermocompression bonded by directly coating the conductive film 51A with the Ag film, Au film, or solder film, and the chip can be fixed via a brazing material such as solder. . Here, the conductive film may be formed on the uppermost layer of the conductive film laminated in a plurality of layers. For example, on a Cu conductive path 51A, two layers of Ni film and Au film are sequentially deposited, three layers of Ni film, Cu film and solder film are sequentially deposited, Ag film, A Ni film can be formed by sequentially coating two layers. In addition, there are many types of these conductive films and laminated structures, but they are omitted here.

  In this semiconductor device, since the conductive path 51 is supported by the insulating resin 50 that is a sealing resin, a support substrate is not required, and the conductive path 51, the semiconductor element 52, the circuit element 52B, and the insulating resin 50 are included. The This configuration is a feature of the present invention. As described in the section of the prior art, since the conductive path of the conventional semiconductor device is supported by the support substrate or supported by the lead frame, a configuration that may be unnecessary is added. However, this semiconductor device is configured with the minimum necessary components and does not require a support substrate, and thus has a feature of being thin and inexpensive.

  In addition to the above-described configuration, the insulating resin 50 covers the semiconductor element 52A and the circuit element 52B and is filled in the separation groove 54 between the conductive paths 52 and integrally supported.

  The space between the conductive paths 51 becomes a separation groove 54, and the insulating resin 50 is filled therewith, so that there is an advantage that mutual insulation can be achieved.

  The insulating resin 50 covers the semiconductor element 52A and the circuit element 52B, fills the separation groove 54 between the conductive paths 51, and exposes the back surface of the conductive path 51 to support it integrally.

  The point that the back surface of the conductive path is exposed is one of the features of the present invention. The back surface of the conductive path can be used for connection to the outside, and the through hole TH having the conventional structure as shown in FIG. 25 can be made unnecessary.

  In addition, when the semiconductor element is directly fixed to the first conductive path 51A through a conductive film made of brazing material, Au, Ag, or the like, the back surface of the first conductive path 51A is exposed, and thus generated from the semiconductor element 52A. The heat to be transmitted can be transmitted to the mounting substrate via the first conductive path 51A. In particular, it is effective for a semiconductor chip that can improve characteristics such as an increase in driving current by heat radiation.

  The back surface of the semiconductor element 52A is set lower than the surfaces of the conductive paths 51B and 51C other than the conductive path 51A to which the semiconductor element 52A is fixed. By doing so, the surface of the semiconductor element 52A can be lowered, and the top of the thin metal wire 55A connecting the semiconductor element 52A and the conductive path 51B can be set low. In the drawing, since the circuit element 52B is higher than the top of the thin metal wire 55A, the thickness of the insulating resin 50 is determined by the circuit element 52B. However, the thickness of the circuit element 52B is thin, and the top of the thin metal wire 55A is When it is higher than the circuit element 52B, the thickness of the insulating resin 50 is determined by the thin metal wire 55A. Therefore, in this case, the thickness of the semiconductor device 53 can be reduced as much as the top of the thin metal wire 55A is lowered.

On the other hand, FIG. 1C shows the back surface polished until the conductive film 55C formed on the back surface of the semiconductor element 52A is exposed. In this case, the semiconductor device has a structure in which the back surface of the separation groove 54 and the back surface of the conductive path 51 substantially coincide. This structure has a feature that the semiconductor device 53 can be moved horizontally as it is because the steps of the back electrodes 10 and 11 shown in FIG. 25 are not provided.

Second Embodiment Explaining Semiconductor Device Next, the semiconductor device 53 shown in FIG. 2 will be explained.

  In this structure, wirings L1 to L3 are formed as the conductive path 51, and the other structure is substantially the same as that of FIG. Therefore, the wirings L1 to L3 will be described.

  As described above, the IC circuit ranges from a small circuit to a large circuit. However, a small circuit is shown in FIG. 2A for convenience of drawing. This circuit is frequently used in an audio amplifier circuit, and is a circuit in which a differential amplifier circuit and a current mirror circuit are connected. As shown in FIG. 2A, the differential amplifier circuit is composed of TR1 and TR2, and the current mirror circuit is mainly composed of TR3 and TR4.

  2B is a plan view when the circuit of FIG. 2A is realized as the semiconductor device, FIG. 2C is a cross-sectional view taken along line AA in FIG. 2B, and FIG. 2D is taken along line BB. It is sectional drawing. A die pad 51A on which TR1 and TR3 are mounted is provided on the left side of FIG. 2B, and a die pad 51D on which TR2 and TR4 are mounted is provided on the right side. External connection electrodes 51B and 51E to 51G are provided on the upper side of the die pads 51A and 51D, and 51C and 51H to 51J are provided on the lower side. B and E indicate a base electrode and an emitter electrode. Since the emitter of TR1 and the emitter of TR2 are connected in common, the wiring L2 is formed integrally with the electrodes 51E and 51G. Since the base of TR3 and the base of TR4 and the emitter of TR3 and the emitter of TR4 are connected in common, the wiring L1 is provided integrally with the electrodes 51C and 55J, and the wiring L3 is integrated with the electrodes 55H and 55I. Is provided.

  The wirings L1 to L3 have characteristics, and if described with reference to FIG. 28, the wirings 25 and 29 correspond thereto. This wiring varies depending on the degree of integration of the circuit device, but the width is as narrow as 25 μm. The width of 25 μm is a numerical value when wet etching is employed, and this width can be further reduced by employing dry etching.

  As is clear from FIG. 2D, since the conductive path 51K constituting the wiring L1 is embedded in the insulating resin 50, the wiring is simply attached to the support substrate as shown in FIGS. Unlike this, it is possible to prevent the wiring 51K from coming off and warping. In particular, as will be apparent from the manufacturing method described later, the anchor effect occurs due to the side surface of the wiring 51K being rough or the eaves being formed on the surface, and the conductive path is disconnected from the insulating resin. No structure. The structure having eaves will be described with reference to FIG.

Since the external connection electrodes 51B, 51C, and 551E to 51J are embedded with the insulating resin as described above, the external connection electrodes 51B, 51C, and 551E to 51J have a structure that does not easily peel off even when an external force is applied from the external lead fixed thereto. Here, the resistor R1 and the capacitor C1 are omitted, but they may be mounted on a conductive path. Further, as will be described later in the embodiment of the mounting structure, it may be mounted on the back surface of the circuit device or on the mounting substrate side.

Third Embodiment Explaining Semiconductor Device Next, the semiconductor device 56 shown in FIG. 8 will be explained.

  In this structure, a conductive film 57 is formed on the surfaces of the conductive paths 51B and 51C, and the other structure is substantially the same as the structure in FIG. Therefore, the conductive film 57 will be described.

  The first feature is that a conductive coating 57 is provided to prevent warping of the conductive path and the semiconductor device.

  In general, due to a difference in thermal expansion coefficient between an insulating resin and a conductive path material (hereinafter referred to as a first material), the semiconductor device itself is warped, and the conductive path is curved or peeled off. Further, since the thermal conductivity of the conductive path 51 is superior to that of the insulating resin, the conductive path 51 first rises in temperature and expands. Therefore, by covering the second material having a smaller thermal expansion coefficient than that of the first material, warping and peeling of the conductive path and warping of the semiconductor device can be prevented. In particular, when Cu is employed as the first material, Au, Ni, Pt, or the like is preferable as the second material. The expansion coefficient of Cu is 16.7 × 10 −6 (minus the sixth power of 10), Au is 14 × 10 −6, Ni is 12.8 × 10 −6, and Pt is 8.9 × 10 6 -6.

  The second feature is that the anchor effect is provided by the second material. Since the eaves 58 are formed of the second material, and the eaves 58 attached to the conductive path 51 are embedded in the insulating resin 50, an anchor effect is generated, and the conductive paths 51B and 51C can be prevented from coming off. It becomes.

As described above, the semiconductor device in which the transistor chip 52A and the passive element 52B as the circuit element are mounted has been described as the semiconductor device. However, the present invention is a semiconductor device in which one semiconductor chip is sealed, as shown in FIG. As described above, the present invention can also be implemented in a semiconductor device in which face-down type elements such as CSP are mounted, or a semiconductor device in which passive elements such as chip resistors and chip capacitors are sealed as shown in FIG. Further, a thin metal wire connected between two conductive paths and sealed may be used. This can be used as a fuse.

First Embodiment Explaining Method of Manufacturing Semiconductor Device Next, a method of manufacturing a semiconductor device 53 will be described with reference to FIGS. 3 to 7 and FIG.

  First, as shown in FIG. 3, a sheet-like conductive foil 60 is prepared. The conductive foil 60 is selected in consideration of the adhesiveness, bonding property, and plating property of the brazing material. As the material, a conductive foil mainly composed of Cu, a conductive foil mainly composed of Al, or Fe is used. A conductive foil made of an alloy such as Ni is employed.

  The thickness of the conductive foil is preferably about 10 μm to 300 μm in consideration of the later etching, and here, a copper foil of 70 μm (2 ounces) is employed. However, basically 300 μm or more and 10 μm or less are good. As will be described later, it is only necessary that the separation groove 61 shallower than the thickness of the conductive foil 60 can be formed.

In addition, the sheet-like conductive foil 60 is prepared by being wound in a roll shape with a predetermined width, and this may be conveyed to each step described later, or a conductive foil cut into a predetermined size is prepared, You may convey to each process mentioned later.

Subsequently, there is a step of removing the conductive foil 60 excluding the region RG in which the semiconductor device is mounted and the region that becomes the conductive paths 51B and 51C other than the region RG, to be thinner than the thickness of the conductive foil 60. Then, there is a step of mounting the semiconductor element 52A in the region RG, and covering the semiconductor element 52A, the separation groove 61 and the conductive foil 60 with the insulating resin 50.

  First, a photoresist (etching resistant mask) PR is formed on the Cu foil 60, and the conductive foil 60 corresponding to the region RG is exposed, and the conductive foil 60 excluding the regions to be the conductive paths 51B and 51C is exposed. In this way, the photoresist PR is patterned (see FIG. 4). Then, etching may be performed through the photoresist PR (see FIG. 5 above).

  The depth of the groove 61 formed by etching is, for example, 50 μm, and the side surface thereof becomes a rough surface, so that the adhesiveness with the insulating resin 50 is improved.

  The side wall of the groove 61 is schematically shown as a straight line, but has a different structure depending on the removal method. This removal process can employ wet etching, dry etching, laser evaporation, and dicing. In the case of wet etching, ferric chloride or cupric chloride is mainly used as the etchant, and the conductive foil is dipped in the etchant or showered with the etchant. Since wet etching is generally non-anisotropic, the side surface has a curved structure.

  In the case of dry etching, etching can be performed anisotropically or non-anisotropically. At present, it is said that Cu cannot be removed by reactive ion etching, but it can be removed by sputtering. Etching can be anisotropic or non-anisotropic depending on sputtering conditions.

  Further, in the laser, the separation groove can be formed by direct laser light irradiation. In this case, the side surface of the separation groove 61 is formed to be straight.

  In dicing, it is impossible to form a complicated bent pattern, but it is possible to form a lattice-like separation groove.

  In FIG. 4, instead of the photoresist, a conductive film resistant to the etching solution may be selectively coated. If the conductive film is selectively deposited on the conductive path, this conductive film becomes an etching protective film, and the separation groove can be etched without employing a resist. Possible materials for this conductive film are Ni, Ag, Au, Pt, Pd, and the like. In addition, these corrosion-resistant conductive films have the feature that they can be used as they are as die pads and bonding pads.

  For example, the Ag coating adheres to Au and also to the brazing material. Therefore, if the Au coating is coated on the back surface of the chip, the chip can be thermocompression bonded to the Ag coating on the conductive path 51 as it is, and the chip can be fixed via a brazing material such as solder. Further, since an Au fine wire can be adhered to the Ag conductive film, wire bonding is also possible. Accordingly, there is an advantage that these conductive films can be used as they are as die pads and bonding pads.

  Subsequently, as shown in FIG. 6, there is a step of mounting the semiconductor element 52A and the circuit element 52B electrically connected to the conductive foil 60 in which the separation groove 61 is formed.

  The semiconductor element 52A is a semiconductor element such as a transistor, a diode, or an IC chip, and the circuit element 52B is a passive element such as a chip capacitor or a chip resistor. Although the thickness is increased, face-down semiconductor elements such as CSP and BGA can also be mounted.

  Here, a bare transistor chip 52A is die-bonded to a conductive path 51A, and an emitter electrode and a conductive path 51B, and a base electrode and a conductive path 51B are fixed by ball bonding by thermocompression bonding or wedge bonding by ultrasonic waves. It is connected via 55A. Reference numeral 52B denotes a passive element such as a chip capacitor or a chip resistor, which is fixed by a brazing material such as solder or a conductive paste 55B.

  Further, as shown in FIG. 7, there is a step of attaching an insulating resin 50 to the conductive foil 60 and the separation groove 61. This can be realized by transfer molding, injection molding, or dipping. As the resin material, a thermosetting resin such as an epoxy resin can be realized by transfer molding, and a thermoplastic resin such as polyimide resin or polyphenylene sulfide can be realized by injection molding.

  In the present embodiment, the thickness of the insulating resin coated on the surface of the conductive foil 60 is adjusted so that, for example, about 100 μm is coated from the top. This thickness can be increased or decreased in consideration of strength.

  The feature of this step is that the conductive foil 60 that becomes the conductive path 51 becomes the support substrate until the insulating resin 50 is coated. Conventionally, as shown in FIG. 26, the conductive paths 7 to 11 are formed by using the support substrate 5 that is not originally required, but in the present invention, the conductive foil 60 that becomes the support substrate is necessary as an electrode material. Material. Therefore, there is a merit that the work can be performed with the constituent materials omitted as much as possible, and the cost can be reduced.

  Further, since the separation groove 61 is formed shallower than the thickness of the conductive foil, the conductive foil 60 is not individually separated as the conductive path 51. Therefore, the sheet-like conductive foil 60 can be handled as a unit, and when the insulating resin is molded, it has a feature that it is very easy to carry to the mold and mount to the mold.

  Subsequently, there is a step of chemically and / or physically removing the back surface of the conductive foil 60 and separating it as the conductive path 51. Here, this removal step is performed by polishing, grinding, etching, laser metal evaporation, or the like.

  In FIG. 7, when an etching resistant mask PR is formed on the conductive foil 60 corresponding to the conductive paths 51A to 51C and the conductive foil exposed from the etching resistant mask PR is etched, the shape shown in FIG. 1A is obtained.

  In FIG. 7, the entire surface is cut by about 30 μm by a polishing apparatus or a grinding apparatus, and the insulating resin 50 is exposed from the separation groove 61. FIG. 1C shows what is realized by this method. As a result, the conductive path 51 having a thickness of about 40 μm is separated. Alternatively, wet etching may be performed on the entire surface of the conductive foil 60 until the insulating resin 50 is exposed, and then the entire surface may be shaved by a polishing or grinding apparatus to expose the insulating resin 50.

As a result, the surface of the conductive path 51 is exposed to the insulating resin 50. Then, the separation groove 61 is shaved to form the separation groove 54 in FIG. (See Figure 7 above)
Finally, a conductive material such as solder is deposited on the back surface of the conductive path 51 exposed if necessary, and the semiconductor device is completed.

  When a conductive film is applied to the back surface of the conductive path 51, a conductive film may be formed in advance on the back surface of the conductive foil. In this case, a portion corresponding to the conductive path may be selectively attached. The deposition method is, for example, plating. The conductive film is preferably made of a material (Ag, Au) that is resistant to etching. Further, when this conductive film is employed, the conductive path 51 can be separated only by etching without polishing.

  In this manufacturing method, only the semiconductor element and the passive element are mounted on the conductive foil 60, but this may be arranged in a matrix form as one unit, or one of the semiconductor elements may be one unit. May be arranged in a matrix. In this case, it separates with a dicing apparatus so that it may mention later.

  By the above manufacturing method, the semiconductor device 53 in which the conductive paths 51B and 51C are embedded in the insulating resin 50 and the top of the metal thin wire 55A is lowered can be realized.

  The feature of this manufacturing method is that the insulating path 50 can be used as a support substrate to separate the conductive path 51. The insulating resin 50 is a material necessary as a material for embedding the conductive path 51, and does not require an unnecessary support substrate 5 unlike the conventional manufacturing method of FIG. Therefore, it has the characteristics that it can be manufactured with a minimum amount of material and cost can be reduced.

The thickness of the insulating resin from the surface of the conductive path 51 can be adjusted when the insulating resin is attached in the previous step. Accordingly, although the thickness varies depending on the semiconductor element to be mounted, the thickness of the semiconductor device 56 has a feature that it can be made thicker or thinner. Here, a semiconductor device in which a 40 μm conductive path 51 and a semiconductor element having a reduced height are embedded in an insulating resin 50 having a thickness of 400 μm is obtained. (See Figure 1 above)

Second Embodiment Explaining Method for Manufacturing Semiconductor Device Next, a method for manufacturing a semiconductor device 56 having eaves 58 will be described with reference to FIGS. In addition, since it is substantially the same as 1st Embodiment except the 2nd material 70 used as eaves being adhere | attached, detailed description is abbreviate | omitted.

  First, as shown in FIG. 9, a conductive foil 60 in which a second material 70 having a low etching rate is coated on a conductive foil 60 made of a first material is prepared.

  For example, it is preferable to deposit Ni on a Cu foil because Cu and Ni can be etched at once with ferric chloride or cupric chloride, and Ni is formed into eaves 58 due to the difference in etching rate. . The thick solid line is the conductive film 70 made of Ni, and the film thickness is preferably about 1 to 10 μm. Further, the thicker the Ni film, the easier the eaves 58 are formed.

The second material may be coated with a material that can be selectively etched with the first material. In this case, first, the film 58 made of the second material is patterned so as to cover the formation region of the conductive path 51, and the conductive foil 60 made of the first material is etched using this film as a mask to form the eaves 58. Because. As the second material, in addition to Ni, Al, Ag, Au, and the like are conceivable. When Ag or Au is used as the second material, partial plating may be formed. (See Figure 9 above)
Subsequently, there is a step of removing the conductive foil 60 excluding the region RG in which the semiconductor element 52A is formed, at least the region to be the conductive paths 51B and 51C, to be thinner than the thickness of the conductive foil 60.

  A photoresist PR is formed on the Ni 70, and the photoresist PR is patterned so that the Ni 70 excluding the region RG in which the semiconductor element 52A is formed and the regions to be the conductive paths 51B and 51C are exposed, and etching is performed through the photoresist. do it.

  As described above, when etching is performed using ferric chloride, cupric chloride etchant, etc., the etching rate of Ni 70 is smaller than the etching rate of Cu 60, and thus eaves 58 appears as etching progresses.

The step of mounting the semiconductor element 52A and the circuit element 52B on the conductive foil 60 in which the separation groove 61 is formed (FIG. 12), the insulating foil 50 is coated on the conductive foil 60 and the separation groove 61, and the conductive foil 60 The process of removing the back surface of the substrate chemically and / or physically and separating it as the conductive path 51 (FIG. 13), and the process up to completion by forming a conductive film on the back surface of the conductive path (FIG. 8) Since it is the same, the description is abbreviate | omitted.

Third Embodiment Explaining Manufacturing Method of Semiconductor Element Next, a manufacturing method in which the semiconductor devices 53 of FIG. 2 are arranged in a matrix and individually separated after sealing will be described with reference to FIGS. . Since this manufacturing method is almost the same as that of the first embodiment, the same parts will be described briefly.

  First, as shown in FIG. 14, a sheet-like conductive foil 60 is prepared.

  In addition, the sheet-like conductive foil 60 is prepared by being wound in a roll shape with a predetermined width, and this may be conveyed to each step described later, or a conductive foil cut into a predetermined size is prepared, You may convey to each process mentioned later.

  Subsequently, a region where the semiconductor element 52A is formed (here, since two semiconductor elements are mounted on the conductive path 51A (51D), it becomes a region where the conductive path 51A (51D) is formed), at least the conductive path There is a step of removing the conductive foil 60 excluding the regions to be 51B, 51C, 51E to 55J thinner than the thickness of the conductive foil 60.

  As shown in FIG. 15, an etching resistant mask PR is formed on the Cu foil 60, and as described above, the conductive regions excluding the region where the semiconductor element 52 </ b> A is formed, at least the regions serving as the conductive paths 51 </ b> B, 51 </ b> C, 51 </ b> E to 55 </ b> J. The photoresist PR is patterned so that the foil 60 is exposed. Then, as shown in FIG. 16, etching may be performed through the photoresist PR.

  The depth of the separation groove 61 formed by etching is, for example, 50 μm, and its side surface is a rough surface, so that the adhesiveness with the insulating resin 50 is improved.

Further, the side walls of the separation groove 61 are schematically illustrated as straight, but have different structures depending on the removal method. This removal process can employ wet etching, dry etching, laser evaporation, and dicing. (For details, refer to the first embodiment)
In FIG. 15, instead of the photoresist PR, a conductive film resistant to the etching solution may be selectively coated. If the conductive film is selectively deposited on the conductive path, this conductive film becomes an etching protective film, and the separation groove can be etched without employing a resist.

  Subsequently, as shown in FIG. 17, the semiconductor element 52A is fixed and electrically connected to the conductive foil 60 corresponding to the RG, and the fine metal wire 55A connected to the electrode on the surface of the semiconductor element 52A is connected to the conductive path 55B. There is a process.

  The semiconductor element 52A is a semiconductor element such as a transistor, a diode, or an IC chip. In addition, a passive element such as a chip capacitor or a chip resistor may be mounted as shown in FIG.

  Here, the bare transistor chip 52A is die-bonded to the groove, and the emitter electrode and the conductive path 51B, and the base electrode and the conductive path 51C are connected via the thin metal wire 55A.

  Furthermore, as shown in FIG. 18, there is a step of attaching an insulating resin 50 to the conductive foil 60 and the separation groove 61. This can be realized by transfer molding, injection molding, or dipping.

  In the present embodiment, the thickness of the insulating resin coated on the surface of the conductive foil 60 is adjusted so as to cover about 100 μm from the top of the semiconductor element. This thickness can be increased or decreased in consideration of strength.

  The feature of this step is that when the insulating resin 50 is coated, the conductive foil 60 that becomes the conductive path 51 becomes a support substrate. Conventionally, as shown in FIG. 26, the conductive paths 7 to 11 are formed by using the support substrate 5 that is not originally required, but in the present invention, the conductive foil 60 that becomes the support substrate is necessary as an electrode material. Material. Therefore, there is a merit that the work can be performed with the constituent materials omitted as much as possible, and the cost can be reduced.

  Further, since the separation groove 61 is formed shallower than the thickness of the conductive foil, the conductive foil 60 is not individually separated as the conductive path 51. Therefore, the sheet-like conductive foil 60 can be handled as a unit, and when the insulating resin is molded, it has a feature that it is very easy to carry to the mold and mount to the mold.

  Subsequently, there is a step of chemically and / or physically removing the back surface of the conductive foil 60 and separating it as the conductive path 51. Here, the removing step is performed by polishing, grinding, etching, metal evaporation of laser, or the like.

  In the experiment, the entire surface is shaved by about 30 μm with a polishing apparatus or a grinding apparatus to expose the insulating resin 50. As a result, the conductive path 51 having a thickness of about 40 μm is separated. Alternatively, wet etching may be performed on the entire surface of the conductive foil 60 until the insulating resin 50 is exposed, and then the entire surface may be shaved by a polishing or grinding apparatus to expose the insulating resin 50.

  As a result, the surface of the conductive path 51 is exposed to the insulating resin 50.

  Further, as shown in FIG. 19, a conductive material CM such as solder is applied to the exposed conductive path 51.

  Finally, as shown in FIG. 20, there is a step of separating each semiconductor device.

  The separation line is indicated by an arrow, and can be realized by dicing, cutting, pressing, chocolate breaking, or the like. In addition, when employ | adopting a chocolate break, what is necessary is just to form a protrusion part in a metal mold | die so that a groove | channel may enter a separation line when coat | covering insulating resin.

  In particular, dicing is often used in a normal method for manufacturing a semiconductor device, and it is preferable because a very small size can be separated.

On the right side of FIG. 27, a flow summarizing the present invention is shown. A semiconductor device can be realized by nine steps of preparation of Cu foil, plating of Ag or Ni, half etching, die bonding, wire bonding, transfer molding, backside Cu foil removal, backside treatment of conductive paths, and dicing. Moreover, all processes can be performed in-house without supplying a support substrate from the manufacturer.

Embodiments for explaining types of semiconductor devices and mounting methods thereof.

  FIG. 21 shows a semiconductor device 81 on which a face-down type semiconductor element 80 is mounted. The semiconductor element 80 corresponds to a bare semiconductor chip, a CSP or BGA whose surface is sealed, and the like. FIG. 22 shows a semiconductor device 83 on which a passive element 82 such as a chip resistor or a chip resistor is mounted. Since these are thin because they do not require a support substrate and are sealed with an insulating resin, they have excellent environmental resistance.

  FIG. 23A explains a real layer structure. The semiconductor devices 53, 81 and 83 of the present invention described so far are mounted on the conductive path 85 formed on the mounting substrate 84 such as a printed board, a metal substrate, or a ceramic substrate.

  In particular, since the conductive path 51A to which the back surface of the semiconductor chip 52 is fixed is thermally coupled to the conductive path 85 of the mounting substrate 84, heat can be radiated through the conductive path 85. Further, when a metal substrate is employed as the mounting substrate 84, the temperature of the semiconductor chip 52 can be further lowered by helping the heat dissipation of the metal substrate. Therefore, the driving capability of the semiconductor chip can be improved.

  For example, power MOS, IGBT, SIT, a transistor for driving a large current, an IC (MOS type, BIP type, Bi-CMOS type) memory element for driving a large current are suitable.

  As the metal substrate, an Al substrate, a Cu substrate, or an Fe substrate is preferable, and an insulating resin and / or an oxide film or the like is formed in consideration of a short circuit with the conductive path 85.

  FIG. 23B shows the semiconductor device used as a mounting substrate. It is as if an element is mounted on a printed circuit board. Since the conductive path is exposed in the semiconductor device 90, an element can be mounted thereon. Here, a chip capacitor 92 and a discrete semiconductor device 91 formed by this manufacturing method are mounted.

It is a figure explaining the circuit apparatus of this invention. It is a figure explaining the circuit apparatus of this invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the circuit apparatus of this invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the manufacturing method of the circuit device of the present invention. It is a figure explaining the circuit apparatus of this invention. It is a figure explaining the circuit apparatus of this invention. It is a figure explaining the mounting method of the circuit apparatus of this invention. It is a figure explaining the mounting structure of the conventional circuit device. It is a figure explaining the conventional circuit device. It is a figure explaining the manufacturing method of the conventional circuit device. It is a figure explaining the manufacturing method of the circuit device of the past and this invention. It is a pattern diagram of an IC circuit applied to the circuit device of the prior art and the present invention.

Explanation of symbols

50 Insulating resin
51A-51C Conductive path
52A Semiconductor device
52B Passive element
53 Semiconductor device
54 Separation groove
58 Eaves

Claims (1)

Preparing a conductive foil;
Forming a groove shallower than the thickness of the conductive foil in the conductive foil corresponding to the region where the semiconductor element is disposed between the conductive paths to be formed; and
Fixing the semiconductor element via an insulating adhesive provided in the groove, and electrically connecting the electrode of the semiconductor element and the desired conductive path by a thin metal wire;
Covering the semiconductor element and the fine metal wire, and molding with an insulating resin so as to fill the groove;
A step of electrically separating the conductive paths from each other by removing the conductive foil from the back surface of the conductive foil .
The semiconductor device is characterized in that the conductive paths and the semiconductor elements corresponding to the semiconductor device are arranged in a matrix and are separated for each semiconductor device after the step of electrically separating the conductive paths. Manufacturing method.

Images