JP4849136B2 - Mounting method of conductive balls - Google Patents

Description

  The present invention relates to a microball mounter for mounting a microball on a surface-mount type semiconductor device such as a BGA or CSP package, and more particularly to a method of mounting a microball on a terminal region of a substrate using a transfer mask.

  With the widespread use of cellular phones, portable computers, and other small electronic devices, there is an increasing demand for miniaturization and thinning of semiconductor devices mounted thereon. BGA packages and CSP packages have been developed and put into practical use in order to meet these requirements.

  A BGA or CSP package is a semiconductor device for surface mounting, and microballs for external connection terminals are mounted on one surface of a substrate of the package and connected thereto. As a method for mounting such a microball, there are a method using a suction head and a method using a transfer mask.

  In the former method, as shown in FIG. 9A, a substrate 3 on which a semiconductor chip 1 is molded with a resin 2 is placed on a stage. A flux or solder paste is formed on the terminal region (electrode land) 4 of the substrate 3. The suction head 5 is formed with suction holes 7 for vacuum-sucking microballs (solder balls) 6, and the suction head 5 is moved toward the substrate 3 while the microballs 6 are sucked. The microball 6 is pressurized and the microball 6 is mounted on the terminal region 4. Such a microball mounting method is disclosed in, for example, Patent Document 1 and Patent Document 2.

  In the latter method, as shown in FIG. 9B, the transfer mask 10 is disposed so as to face the substrate 3. A through-hole 11 having the same pattern as the pattern of the terminal region 4 of the substrate 3 is formed in the transfer mask 10. The microball 6 supplied onto the transfer mask 10 is dropped into the through hole 11 and mounted on the terminal region 4. Thereafter, the microball 6 and the terminal region 4 are bonded to each other by reflow to form a bump electrode for external connection of a BGA or CSP package. Such a microball mounting method is disclosed in Patent Document 3, for example.

JP 2001-332899 A Japanese Patent Laid-Open No. 8-335771 JP 2004-327536 A

  A conventional microball mounting apparatus using a transfer mask has the following problems. As shown in FIG. 10A, the substrate 3 has the resin 2 placed on the backup plate 20, and the transfer mask 10 has its end 10a fixed to a fixed base (not shown). Are arranged opposite to each other. A magnet 22 is provided below the backup plate 20, and the transfer mask 10 including a metal mask is attracted to the backup plate 20 side by the magnetic force of the magnet 22, and the transfer mask 10 is in close contact with the substrate 3. In this state, the microball 6 is dropped onto the terminal region 4 through the through hole 11.

  After the microball 6 is mounted, as shown in FIG. 10B, the backup plate 20 and the magnet 22 are lowered, and the substrate 3 is released from the transfer mask 10. At this time, due to the action of the magnetic force of the magnet 22, the central portion 10b of the transfer mask 10 to which the end portion 10a is fixed bends downward in a concave shape, and the release of the transfer mask 10 is delayed as much as the central portion. When the transfer mask 10 is bent, the position of the through hole 111 of the transfer mask 10 in the vertical direction is shifted, so that the position of the microball 6 mounted on the substrate is shifted by the through hole 11.

  Further, as shown in FIG. 11A, the transfer mask 10 includes a metal mask 30 and a resin layer 32 formed under the metal mask 30, and the metal mask 30 penetrates somewhat larger than the diameter of the microball 6. A hole 11 is formed, and an opening 34 larger than the diameter of the through hole 11 is formed in the resin layer 32. After the microball 6 is mounted on the terminal region 4 of the substrate 3, when the substrate 3 is released from the transfer mask 10, as described above, the transfer mask 10 is bent or the accuracy of the through hole 11 is poor. Then, as shown in FIG. 11B, there is a problem that the microball 6 comes into contact with the side surface of the through hole 11 and stays there, and the microball 6 is not successfully dropped onto the terminal region 4.

  Such misalignment and mounting failure of the microballs cause defects and defects in the semiconductor device, resulting in a decrease in yield and an increase in manufacturing cost.

The present invention solves such a conventional problem, and provides a conductive ball mounting apparatus and a mounting method for mounting a conductive ball on a terminal area of a substrate more accurately and reliably. The purpose is to do.
Still another object of the present invention is to provide a conductive ball mounting device and a mounting method thereof that can improve the yield of a surface-mount type semiconductor device such as BGA or CSP and reduce the manufacturing cost. .

A method for mounting a conductive ball according to the present invention is a method for mounting a conductive ball on a plurality of terminal regions formed on one surface of a substrate, wherein a plurality of through holes corresponding to the plurality of terminal regions of the substrate are formed. A step of arranging the substrate so as to face the mask, and a step of attracting the mask to the mounting position by magnetic force, wherein the attractive force of the mask at the center of the substrate is smaller than the attractive force of the mask at the peripheral portion of the substrate gives said step, dropped into a plurality of through-holes of the conductive balls on a substrate, a step of mounting the conductive balls on the corresponding terminal region, viewed including the step of applying vibration to the mask, the vibration in the mask The process is performed when the substrate is separated from the mask by a certain distance .

Preferably , the vibration includes at least one horizontal impact on the mask.

  According to the present invention, when the substrate is released from the mask, the attractive force of the mask at the central portion of the substrate is made smaller than the attractive force of the mask at the peripheral portion of the substrate. The bending is suppressed, and the conductive ball mounted on the substrate is prevented from being displaced by the mask. Further, when the substrate is released from the mask, the conductive ball remaining in the through hole of the mask can be dropped onto the terminal region of the substrate by applying vibration to the mask. Thereby, the yield of the semiconductor device can be improved and the manufacturing cost can be reduced.

It is a schematic flow which shows the manufacturing process of the semiconductor device which concerns on the Example of this invention. FIG. 2A is a plan view illustrating a substrate on which a semiconductor chip is mounted, and FIG. 2B is a cross-sectional view when one semiconductor chip on the substrate is molded with resin. It is typical sectional drawing which shows schematic structure of the microball mounting apparatus which concerns on the Example of this invention. It is a top view of a transfer mask. FIG. 5A is a diagram in which the planar sizes of the transfer mask, the substrate, and the magnet unit are projected in the vertical direction, and FIG. 5B is a diagram schematically showing these sizes from the side. It is typical sectional drawing which shows a state when mounting a board | substrate in a microball mounting apparatus. FIG. 7A shows a state in which vibration is applied to the transfer mask, and FIG. 7B shows a state in which the microball is shaken off by the vibration. It is a figure which shows the modification of the magnet part of a present Example. FIG. 9A shows a conventional microball mounting method. FIG. 9A shows a mounting method using a suction head, and FIG. 9B shows a mounting method using a transfer mask. It is a figure explaining the subject by the conventional transfer mask. It is a figure explaining the subject by the conventional transfer mask.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a schematic flow chart of a preferable manufacturing process of a semiconductor device according to an embodiment of the present invention. In this embodiment, a BGA package is taken as an example of a semiconductor device for surface mounting. First, a plurality of semiconductor chips are mounted on a substrate (step S101), and the mounted semiconductor chips are molded with resin (step S102). Next, a microball is mounted on the terminal region (electrode land) of the substrate (step S103), the microball and the terminal region are metal-bonded by reflow (step S104), and the substrate is cut for each semiconductor chip. Is performed (step S105).

  FIG. 2A is a plan view of a substrate on which a semiconductor chip is mounted, and FIG. 2B is a view showing a cross section of one semiconductor chip mounted on the substrate. The configuration of the substrate 100 is not particularly limited, but a multilayer wiring substrate or a film substrate in which an insulating layer and a wiring layer are stacked can be used. As for the external shape of the board | substrate 100, a longitudinal direction is about 230 mm and a transversal direction is about 62 mm, for example. On the surface of the substrate 100, the semiconductor chips 102 are mounted in a two-dimensional array. For example, a 5 × 5 semiconductor chip is used as one block 104 A, and such blocks 104 B, 104 C, and 104 D are arranged in the longitudinal direction of the substrate 100.

  As shown in FIG. 2B, one semiconductor chip 102 is fixed on the substrate 100 via an adhesive such as die attach, and the electrodes formed on the surface of the semiconductor chip 102 are bonded to the substrate 100 by bonding wires 106. Connected to the upper wiring pattern. Alternatively, the semiconductor chip 102 may be flip chip mounting in which the bump electrodes formed on the surface thereof are faced down and connected to the wiring pattern of the substrate. The wiring pattern formed on the front surface of the substrate 100 is electrically connected to a plurality of terminal regions (portions shown in black in the figure) 108 formed in an array on the back surface of the substrate 100 via internal wiring. Is done. As will be described later, the terminal region 108 provides a region for connecting microballs for external connection terminals of a BGA package. For example, several tens to several hundreds of microballs can be connected to one BGA package. .

  Each semiconductor chip 102 on the substrate 100 is molded with a resin 110. In this embodiment, one block made up of 5 × 5 semiconductor chips is molded together. However, each of the semiconductor chips 102 may be individually molded. For example, the height of the resin 110 from the surface of the substrate 100 is about 450 microns, and the thickness of the substrate 100 is about 240 microns.

  Next, the microball mounting device will be described. FIG. 3 is a diagram schematically showing a cross-sectional structure of the microball mounting apparatus. Here, for easy understanding of the description, a substrate on which one semiconductor chip is mounted and a transfer mask corresponding thereto are illustrated. Therefore, it is not necessarily the same scale as the actual microball mounting device.

  The microball mounting apparatus 200 includes a transfer mask 210 in which a plurality of through holes having the same pattern as a pattern of a plurality of terminal regions 108 of the substrate 100 and a metal such as stainless steel are formed. The backup plate 220 to be placed, the fixing block 230 which is disposed around the backup plate 220 and fixes the end portion 210a of the transfer mask 210 by vacuum suction or the like, and the lower portion of the backup plate 230. A magnetic part 240 attracted by the motor, a drive device 250 that moves the backup plate 230 and the magnet part 240 in the vertical direction, and a vibration device 260 that vibrates the end 210a of the transfer mask 210 fixed to the fixed block 230. Yes.

  FIG. 4 shows a plan view of the transfer mask and enlarges the mask area corresponding to one semiconductor chip. In the transfer mask 210, through holes 216 are formed in regions corresponding to the blocks 104 </ b> A to 104 </ b> D of the substrate 100. As shown in FIG. 3, the transfer mask 210 is formed by aligning two layers of masks. The first layer mask 212 is made of a metal such as stainless steel, and a through hole 216 is formed in the metal mask. Yes. The second layer mask 214 is made of a resin layer such as acrylic or epoxy resin, and has a rectangular opening 218 corresponding to a terminal region formed on a substrate of one semiconductor chip. Taking FIG. 4 as an example, a 5 × 5 through-hole 216 is formed in the first layer mask 212 corresponding to the terminal region (5 × 5) formed on the substrate of one semiconductor chip. In the layer mask 214, a rectangular opening 218 including a region of a 5 × 5 through hole 216 is formed.

  The transfer mask 210 has an end 210a that forms the outer periphery thereof fixed by a fixing block 230, and the central portion of the transfer mask 210 can be bent in the vertical direction with the end 210a as a fulcrum. As will be described later, the center portion of the transfer mask 210 is attracted to the backup play 220 side by the magnetic force of the magnet portion 240 when the microball is mounted, and the transfer mask 210 is in close contact with the microball mounting surface of the substrate 100. .

  In the present embodiment, the magnet part 240 includes two permanent magnets having relatively different magnetic forces. FIG. 5A is a diagram in which the planar sizes of the transfer mask, the substrate, and the magnet portion are projected in the vertical direction, and FIG. 5B is a diagram schematically showing these sizes from the side. . In the same figure, P1 shows the edge (outer edge) of the transfer mask 210, P2 shows the boundary of the edge part 210a of the transfer mask 210 which the fixed block 230 fixes, P3 shows the edge of the board | substrate 100, and the magnet part 240. P4 indicates the boundary of the relatively weak magnetic field at the center of the magnet part 240.

  In the transfer mask 210, an end portion 210a shown from the edge P1 to the boundary P2 is fixed by a fixing block 230. When the substrate 100 is placed on the backup plate 220, the substrate 100 is arranged so as to be line symmetric with respect to the center line C of the transfer mask 210. The magnet unit 240 has substantially the same planar size as the substrate 100, but the magnet region 242 has a relatively strong magnetic force corresponding to the peripheral region 100 a of the substrate 100 and the inner region 100 b inside the peripheral region 100 a of the substrate 100. Corresponding to the magnet region 244 having a relatively weak magnetic force.

  Next, the operation of the microball mounting device will be described. First, as shown in FIG. 6, the backup plate 220 and the magnet unit 240 are lowered to a predetermined position by the driving device 250 so that the microball mounting surface (surface on which the terminal region 108 is formed) of the substrate 100 faces upward. The mold resin surface is placed on the backup plate 220. On the other hand, the end 210 a of the transfer mask 210 is fixed by the fixing block 230.

  Next, the driving device 250 raises the backup plate 220 and the magnet unit 240 to a certain position. For example, the backup plate 220 is raised to a position where the microball mounting surface of the substrate contacts the second layer mask 214 of the transfer mask 210. Thereby, the transfer mask 210 is attracted to the backup plate side by the magnetic force of the magnet part 240, and the transfer mask 210 presses the microball mounting surface of the substrate. When the substrate 100 is warped, the warp of the substrate is corrected by pressing with the transfer mask 210.

  Next, as shown in FIG. 3, the microballs 270 are supplied to the surface of the transfer mask 210. The microball 270 is, for example, a metal ball in which a solder layer is formed on the surface of a core made of a metal such as copper or a resin, and its diameter corresponds to 300 microns, 180 microns, or fine pitch. About 100 microns. The diameter of the through hole 216 is about 10 to 20 microns larger than the diameter of the microball 270. Here, the thickness of the transfer mask 210 is preferably about 1.1 times that of the microball 270, and the thickness of the resin layer 214 of the transfer mask 210 is selected from the viewpoint of preventing the lateral displacement of the microball. About 1/3 of this is preferable. For example, when the diameter of the microball 270 is 180 microns, the thickness of the metal mask 212 is about 140 microns, the thickness of the resin layer 140 is 60 microns, and the thickness of the transfer mask 210 is about 200 microns. .

  The supplied microball 270 is dropped into the through hole 216, and the microball 270 is mounted on the terminal region 108 of the substrate 100. The terminal region 108 is preferably formed with a flux or a solder paste to adsorb the microballs 270.

  When the mounting of the microball 270 is completed, the driving device 250 lowers the backup plate 220 and the magnet unit 240. Immediately after the descent, the magnetic force by the magnet part 240 acts on the transfer mask 210, but since the magnetic force of the magnet area 244 in the central part of the magnet part 240 is smaller than the magnetic force of the magnet area 242 in the peripheral part, the transfer mask 210. The magnetic force acting on the central portion of the transfer mask 210 is relaxed compared to the peripheral portion, and the deflection of the central portion of the transfer mask 210 is suppressed. As a result, interference with the microball 270 by the through-hole 216 of the transfer mask 210 is prevented, and the microball is not displaced.

  Further, in this embodiment, in order to more surely mount the microball, the driving device 250 stops the backup plate when the backup plate 220 is lowered by a certain distance, and in this state, the transfer device 260 performs transfer. Vibration is applied to the mask 210. The vibration is preferably horizontal vibration and the vibration may be at least one impact. As a method for applying vibration, known means can be used. For example, a cam is rotated by a motor, and a horizontal vibration or impact is applied to the transfer mask by sliding the cam surface and the side of the transfer mask. it can.

  When the substrate is released from the transfer mask, when the deflection remains in the transfer mask 210 or when the accuracy of the through hole 216 is not good, as shown in FIG. The microball 270 may come into contact with the 216 and remain there. In this embodiment, by applying vibration to the transfer mask 210, the microballs 270 remaining in the through holes 216 are detached from the through holes 216 as shown in FIG. Can be shaken off.

  The position where the backup plate 220 is stopped is preferably about the thickness of the transfer mask 210. This is because if the distance between the transfer mask 210 and the microball mounting surface of the substrate becomes too large, there is a possibility that the microball dropped from the through-hole 216 may not be accurately mounted on the terminal region 108. Furthermore, the distance by which the transfer mask 210 is moved in the horizontal direction by vibration is, for example, ½ or less of the diameter of the microball. This is because the microball 270 is not accurately mounted on the terminal region 108 if the moving distance becomes too large.

  Through the above steps, the substrate on which the microballs are mounted is removed from the mounting device, and then transferred to a reflow process, where the solder layer formed on the microball surface and the terminal region 106 are bonded to each other. Then, the substrate is cut into individual semiconductor chips by a blade, and a BGA package on which microballs 270 are mounted is obtained.

  As described above, according to the present embodiment, the magnetic force acting on the transfer mask is weakened at the central portion rather than the peripheral portion, thereby suppressing the deflection of the transfer mask when the substrate is released from the transfer mask. It is possible to prevent the displacement of the microball due to the interference. Furthermore, by applying vibration to the transfer mask, the microballs remaining in the through holes can be more reliably and accurately mounted on the terminal region of the substrate.

  Next, a modification of the magnet part of the present embodiment will be described. In the above embodiment, two permanent magnets having different magnetic forces are used in order to vary the attractive force to the transfer mask. However, in addition to this, for example, the attractive force may be varied using one permanent magnet. May be. As shown in FIG. 8A, a distance D1 from the magnet region 302 to the transfer mask 210 corresponding to the peripheral region 100a of the substrate 100 is defined as a distance D1 from the magnet region 304 to the transfer mask 210 corresponding to the inner region 100b of the substrate 100. The thickness of the permanent magnet 300 can be changed so as to be smaller than the distance D2. Thereby, the attraction | suction by the magnetic force to the transfer mask 210 can be enlarged in a peripheral part, and can be made small in a center part.

  Further, as shown in FIG. 8B, the thickness of the backup plate 222 located between the permanent magnet 310 and the transfer mask 210 is set so that the region 224 corresponding to the peripheral region 100a of the substrate 100 is the inner region 100b of the substrate. You may make it make it thinner than the area | region corresponding to. Thereby, the attracting force by the magnetic force of the transfer mask 210 by the permanent magnet 310 can be strengthened at the peripheral portion and small at the central portion.

  Although the preferred embodiment of the present invention has been described in detail, the present invention is not limited to the specific embodiment according to the present invention, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims. It can be changed.

  The central part and the peripheral part where the attractive force due to the magnetic force of the magnet part 240 is different can be appropriately changed according to the size, shape, rigidity, etc. of the transfer mask, and further, the size, shape, terminal area of the substrate It can change suitably according to the arrangement | sequence etc. of these. In the present invention, what is important is that the attracting force to the transfer mask is relatively smaller at the central portion than at the peripheral portion.

  In the above embodiment, the BGA package is taken as an example, but the present invention can also be applied to a CSP package and other surface mount type semiconductor devices. In the above embodiment, the transfer mask has an example of a two-layer structure including a metal layer and a resist. However, the transfer mask may have a single-layer structure or a stacked structure of three or more layers. May be.

  The conductive ball mounting apparatus according to the present invention is used in a semiconductor manufacturing apparatus for manufacturing a surface mounting type semiconductor device.

100: Substrate 102: Semiconductor chips 104A, 104B, 104C, 104D: Block 106: Bonding wire 108: Terminal region 110: Mold resin 200: Microball mounting device 210: Transfer mask 212: First layer mask 214: Second layer mask 216: Through hole 218: Opening 220: Backup plate 230: Fixed block 240: Magnet part 242, 244: Magnet region 250: Drive device 260: Vibration device 270: Micro ball 300, 310: Permanent magnet 302, 304: Magnet region

Claims (2)

A method of mounting conductive balls on a plurality of terminal regions formed on one surface of a substrate,
Arranging the substrate so as to face a mask in which a plurality of through holes corresponding to a plurality of terminal regions of the substrate are formed;
The step of attracting the mask to the mounting position by magnetic force, wherein the attracting force of the mask at the center of the substrate is smaller than the attracting force of the mask at the peripheral portion of the substrate;
Dropping a conductive ball onto a substrate from a plurality of through holes and mounting the conductive ball on a corresponding terminal region; and
Applying vibration to the mask;
Only including,
The mounting method , wherein the step of applying vibration to the mask is performed when the substrate is separated from the mask by a certain distance . Vibration includes at least one horizontal impact on the mask, mounting method of claim 1.

Images