1. Technical Field
The present disclosure relates to flip attach and, more specifically, to methods of creating molds of variable solder volumes for flip attach.
2. Discussion of the Related Art
In the process of manufacturing electronic equipment, semiconductor devices, such as integrated circuits (ICs) are often encased in a protective package and mounted onto a printed circuit board (PCB) or other electronic device.
Conventionally, semiconductor devices may be mounted onto a PCB using a series of thin wire interconnects. However, as semiconductor devices become smaller and more complex, the wire interconnects must become thinner and closer together. Many modern semiconductor devices are so small and complex that wire interconnects are no longer practical. Accordingly, other methods for chip mounting have been developed.
Flip chip mounting methods are used to mount a semiconductor device without the need for wire connections. In flip chip mounting, bumps of solder are formed on the chip's connection pads during wafer processing. The chip may then be inverted such that the solder bumps directly contact the PCB or other associated external circuitry. Then, in a process called controlled collapse chip connection (C4), the solder bumps are reflowed and electrical connection is achieved. Electrically-insulating adhesive may then be used to underfill the space between the chip and the PCB to provide a stronger mechanical connection.
Solder may be applied to a semiconductor chip to form interconnects. Methods for applying the solder bumps to the chip have been developed. For example, solder may be applied by evaporation through a shadow mask, electroplated into a Riston opening, or screen printing. Other approaches include injection molded solder (IMS) and direct solder ball attach.
For example, the surface of the wafer may be screened with solder paste before the chip die is cut. However, the solder paste, which generally includes flux and solder alloy particles, may lack a consistent and uniform composition, especially as the size of the solder bumps decreases to accommodate smaller chips. Particular care may be given to provide for a highly uniform and consistent solder paste, however, such care generally comes at a high cost. Moreover, another problem with using solder paste screening techniques in modern high density devices is the reduced pitch between bumps. Since there is a large reduction in volume from a screened paste to the resulting solder bump, the screen holes must be significantly larger in diameter than the final bumps. Thus stringent dimensional control of the bumps makes the solder paste screening technique impractical for applications in high density devices.
More recently developed injection molded solder (IMS) techniques attempt to solve these problems by dispensing molten solder instead of solder paste. According to these methods, a transfer mold having an array of cavities is filled with injected solder. The mold is then disposed over a semiconductor chip or chip packaging substrate such that the filled cavities align with the points of electrical contact on the chip. A combination of heat and gas pressure is applied to transfer the solder pattern onto the chip. Methods for IMS are described in U.S. Pat. Nos. 5,244,143; 6,056,191; and 6,105,852, the disclosures of which are hereby incorporated by reference in their entirety.
Transfer molds are generally made of glass or polymeric substrates. A masking material may then be deposited on the mold and a pattern of holes may be formed on the mask. The layout of the patterned holes is determined by the footprint of the chip that is to receive the solder bumps. The mask is then etched to form the cavities and the mask is then removed. Because most etch processes are isotropic and have a constant etch rate in all directions, the diameter of the holes in the mask and the spacing between the holes in the mask determine the diameter, pitch and etch depth of the cavities that are formed during etching.
A method for fabricating a solder transfer mold includes masking a substrate with a masking agent. A pattern is transferred to the substrate mask. The masked substrate is etched until cavities of a first volume are formed. The cavities of the first volume are selectively coated. The masked substrate is etched until cavities of a second volume are formed.
A method for fabricating a solder transfer mold includes covering a substrate having anisotropic etching properties with a masking layer. The masking layer is patterned to create a plurality of openings of at least two different sizes. The substrate is etched through the patterned mask to generate a plurality of cavities of at least two different volumes.
A solder mold includes a substrate. The substrate includes a plurality of cavities for holding solder to be transferred to an integrated circuit. The plurality of cavities includes cavities of at least two different volumes.
A method for applying solder bumps directly to an integrated circuit includes filling a plurality of cavities within a solder mold with solder. The solder mold is placed in proximity with the integrated circuit. The solder is transferred from the pluralities of the cavities to the integrated circuit. The solder mold includes a substrate and the plurality of cavities and the plurality of cavities include cavities of at least two different volumes.
A method for fabricating a solder transfer mold having solder cavities of multiple different volumes includes placing multiple alternating layers of a first protective material and a second protective material on a substrate. The following etch steps are repeated: a first protective material etch is performed, a second protective material etch is performed, and a substrate etch is performed. The number of alternating layer pairs is equal to the number of etch step repetitions and is equal to the number of different volumes.
A method for generating a solder mold includes etching a first set of cavities of a first volume in a solder mold substrate. The first set of cavities continue to be etched while etching a second set of cavities of a second volume. The second volume is smaller than the first volume.
A more complete appreciation of the present disclosure and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:
In describing the exemplary embodiments of the present disclosure illustrated in the drawings, specific terminology is employed for sake of clarity. However, the present disclosure is not intended to be limited to the specific terminology so selected, and it is to be understood that each specific element includes all technical equivalents which operate in a similar manner.
Exemplary embodiments of the present invention seek to provide injection molded solder (IMS) techniques that allow for solder bumps of varying volumes within a single transfer mold. A single transfer mold may thereby be used to transfer solder bumps of different diameter and pitch without having to use multiple transfer molds. For example, some transferred solder bumps may have a diameter of 4 thousandths of an inch (mils) and a pitch of 8 mils (4-on-8), while other transferred solder bumps may have a diameter of 3 mils and a pitch of 6 mils (3-on-6), and still other transferred solder bumps may have a diameter of 2 mils and a pitch of 4 mils (2-on-4), etc.
Many available techniques may be used to mask off the fully formed cavities to prevent over-etching and allow for multiple cavities of various sized on a single mold substrate. According to one exemplary embodiment of the present invention, fully formed cavities of the mold may be selectively coated with a metal layer to prevent over-etching. Accordingly, a metal layer may be used as a sacrificial mask to prevent over-etching in fully formed cavities.
According to another exemplary embodiment of the present invention, a polymer may be deposited into fully formed cavities of the mold to prevent over-etching. The polymer may be defined by either standard photolithographic techniques, for example, photo-imageable polymide, or by selective removal using laser ablation.
According to another exemplary embodiment of the present invention, the mold substrate may be made of anisotropic crystalline silicon. When using such a substrate, the process is self-limiting and the diameter of the hole of a single masking layer will automatically determine the final depth of the fully formed cavity. Accordingly, a single mask and a single etching operation may be sufficient to create fully formed varieties of varying sizes and volumes.
Resist removal and wet isotropic mold etching may be performed. Etching may continue until the smallest cavities are fully formed.
The metal mask 24 may be removed and etching may resume until the next-higher volume cavities are fully formed. In
As seen in
Resist removal and wet isotropic mold etching may be performed. Etching may continue until the smallest cavities are fully formed.
Where the selected polymer 34 is curable, the polymer 34 may be cured to prevent over etching when etching continues. Etching may then resume until the next-higher volume cavities are fully formed. In
The completed IMS solder mold may then be used to transfer solder onto a die by filling the cavities with solder and transferring the solder onto the die. Here, known solder transfer processes may be used.
The substrate 44 may comprise, for example, a single silicon crystal with a (100) orientation. First, the crystal substrate may be oxidized to provide an oxidation layer 45. The oxidation layer 45 may be, for example, approximately 5000 Å thick. The oxidation layer 45 may then be patterned, for example, with square or rectangular features. The patterning may be accomplished, for example, using lithographic techniques. The patterned oxidation layer 45 may then be etched, for example, etched in BHF, to open one or more oxide windows 46 and 47 of varying sizes. For example, a smaller oxide window 46 may be 1 mil by 1 mil and a larger oxide window 47 may be 2 mil by 2 mil. Anisotropic wet etching may then be performed on the substrate 44 masked by the patterned oxidation layer 45. The wet etch may be performed, for example, using EPPW or KOH solution or any chemistry suited for anisotropic etching of silicon.
When the side of the square or rectangular features are aligned to the (110) direction of the wafer, the resulting etch cavities are pyramid shaped with four sides following the (111) plane. The resulting pyramid shaped cavities are self-limiting in size. Due to the anisotropic nature of the silicon substrate, the volumes of the resultant cavities are a direct result of the size and shape of the patterned opening. Accordingly, larger openings may result in deeper cavities. The triangles 48 and 49 in
After the cavities have been formed, a final oxidation step may be performed to provide a protective oxide layer (not shown) over the substrate 44. Accordingly, cavities of various volumes may be obtained with a single fabrication process flow.
As seen in
Each copper layer may be approximately 500 to 5000 Angstroms thick. Each chromium layer may be approximately 100-200 Angstroms thick.
As seen in
Again, it is noted that the alternating layers may be of materials other than copper and chromium as long as the layers have different etch sensitivities, however, as described herein, copper and chromium are illustrated to is provide a simple example.
As seen in
Then as seen in
Then, as seen in
Then, as seen in
Then, as seen in
Where there are to be more than two different sized volumes, additional copper/chromium layers may be used and additional copper/chromium/glass etch steps may be performed until all volumes are fully etched to their respective desired volumes.
This process is illustrated in
If all cavities have been fully etched (yes, Step S605) then the process is complete. However, if all cavities have not been fully etched (no, Step S605) then a new mask is applied (Step S601) such that all previously patterned openings are opened again along with the next-largest set of cavities (Step S606). This process continues until all cavities are fully etched. Because of the layered approach, the first glass etch step only etches the largest cavity, the next glass etch step further etches the largest cavity and begins etching the next-largest cavity, additional glass etch steps proceed accordingly until all cavities have been fully etched. The length of time the glass etch steps are performed for are calculated according to the desired etch volume for each cavity.
According to this approach, where there are cavities of two sizes, the larger cavity will be etched twice and the smaller cavity will be etched once, as illustrated above with reference to
Because the top layer may be the copper layer, at some point in the process, for example, after the first copper etch (Step S602), the top copper layer may be thickened. The thickened copper layer may provide enhanced structural support for the film stacks and may minimize the risk of the film stacks collapsing during glass etch.
The above specific exemplary embodiments are illustrative, and many variations can be introduced on these embodiments without departing from the spirit of the disclosure or from the scope of the appended claims. For example, elements and/or features of different exemplary embodiments may be combined with each other and/or substituted for each other within the scope of this disclosure and appended claims.
This application is a continuation of U.S. patent application Ser. No. 11/769,389, filed Jun. 27, 2007, the contents of which are herein incorporated by reference in its entirety.
Number | Date | Country | |
---|---|---|---|
Parent | 11769389 | Jun 2007 | US |
Child | 11844070 | US |