Methods of forming and operating microelectronic devices including dummy chips

This application is a continuation of U.S. patent application Ser. No. 14/730,231, filed Jun. 3, 2015, pending, the disclosure of which is hereby incorporated herein in its entirety by this reference.

The present invention relates generally to the field of semiconductor packaging, and more particularly to a wafer-level package (WLP) with a substrate-less or TSV-less (TSV: Through Substrate Via) interposer and a method for manufacturing the same.

As known in the art, fan-out wafer-level packaging (FOWLP) is a packaging process in which contacts of a semiconductor die are redistributed over a larger area through a redistribution layer (RDL) that is typically formed on a substrate such as a TSV interposer.

The RDL is typically defined by the addition of metal and dielectric layers onto the surface of the wafer to re-route an Input/Output (I/O) layout into a looser pitch footprint. Such redistribution requires thin film polymers such as benzocyclobutene (BCB), polyimide (PI), or other organic polymers and metallization such as Al or Cu to reroute the peripheral pads to an area array configuration.

The TSV interposer is costly because fabricating the interposer substrate with TSVs is a complex process. Thus, forming FOWLP products that include an interposer having a TSV interposer may be undesirable for certain applications.

In wafer-level packaging, the wafer and dies mounted on the wafer are typically covered with a relatively thick layer of molding compound. The thick layer of the molding compound results in increased warping of the packaging due to coefficient of thermal expansion (CTE) mismatch, and the thickness of the packaging. It is known that wafer warpage continues to be a concern.

Warpage can prevent successful assembly of a die-to-wafer stack because of the inability to maintain the coupling of the die and wafer. Warpage issue is serious, especially in a large-sized wafer, and has raised an obstacle to a wafer-level semiconductor packaging process that requires a fine-pitch RDL process. Therefore, there remains a need in the art for an improved method of manufacturing wafer-level packages.

The present invention is directed to provide an improved semiconductor device and fabrication method that is capable of reducing the total used amount of molding compound on an interposer, thereby alleviating post-molding warpage.

In one aspect of the invention, a semiconductor device includes an interposer having a first side and a second side opposite to the first side; at least one active chip mounted on the first side within a chip mounting area through a plurality of first bumps; at least one dummy chip mounted on the first side within a peripheral area being adjacent to the chip mounting area; a molding compound disposed on the first side, the molding compound covering the at least one active chip and the at least one dummy chip; and a plurality of solder bumps mounted on the second side.

According to one embodiment of the invention, the dummy chip is mounted on the first side through a plurality of second bumps disposed on dummy pads within the peripheral area.

According to another embodiment of the invention, the dummy chip is mounted directly on the first side with an adhesive.

These and other objectives of the present invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiments that are illustrated in the various drawings.

The accompanying drawings are included to provide a further understanding of the embodiments, and are incorporated in and constitute a part of this specification. The drawings illustrate some of the embodiments and, together with the description, serve to explain their principles. In the drawings:

FIGS. 1 through 8 are schematic diagrams showing an exemplary method for fabricating a wafer-level package (WLP) with a substrate-less (or TSV-less) interposer according to one embodiment of the invention, wherein,

FIGS. 1 through 7 are schematic, cross-sectional views of the intermediate product during the manufacturing process of the WLP;

FIG. 8 is a top view showing the exemplary layout of the active chips and dummy chips on the RDL; and

FIGS. 9 through 13 are schematic diagrams showing an exemplary method for fabricating a WLP with a substrate-less interposer according to another embodiment of the invention.

In the following detailed description of the invention, reference is made to the accompanying drawings, which form a part hereof, and in which is shown, by way of illustration, specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. Other embodiments maybe utilized and structural changes may be made without departing from the scope of the present invention.

The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims, along with the full scope of equivalents to which such claims are entitled.

One or more implementations of the present invention will now be described with reference to the attached drawings, wherein like reference numerals are used to refer to like elements throughout, and wherein the illustrated structures are not necessarily drawn to scale. The terms “die,” “semiconductor chip,” and “semiconductor die” are used interchangeably throughout the specification.

The terms “wafer” and “substrate” used herein include any structure having an exposed surface onto which a layer is deposited according to the present invention, for example, to form the circuit structure such as a redistribution layer (RDL). The term “substrate” is understood to include semiconductor wafers, but is not limited thereto. The term substrate is also used to refer to semiconductor structures during processing, and may include other layers that have been fabricated thereupon.

With reference to FIGS. 1 through 8 , depicted are schematic diagrams showing an exemplary method for fabricating a wafer-level package (WLP) with a substrate-less (or TSV-less) interposer according to one embodiment of the invention.

As shown in FIG. 1 , a carrier 300 is prepared. The carrier 300 may be a releasable substrate material with an adhesive layer (not explicitly shown). At least a dielectric layer or a passivation layer 310 is then formed on a top surface of the carrier 300 . The passivation layer 310 may comprise organic materials such as polyimide (PI) or inorganic materials such as silicon nitride, silicon oxide, or the like.

As shown in FIG. 2 , subsequently, a redistribution layer (RDL) 410 is formed on the passivation layer 310 . The RDL 410 may comprise at least one dielectric layer 412 and at least one metal layer 414 . The dielectric layer 412 may comprise organic materials such as polyimide (PI) or inorganic materials such as silicon nitride, silicon oxide, or the like, but not limited thereto. The metal layer 414 may comprise aluminum, copper, tungsten, titanium, titanium nitride, or the like.

According to the illustrated embodiment of FIG. 2 , the metal layer 414 may comprise a plurality of bump pads 415 a and dummy pads 415 b exposed from a top surface of the dielectric layer 412 . The bump pads 415 a are disposed within a chip mounting area, while the dummy pads 415 b are disposed outside the chip mounting area such as a peripheral area around the chip mounting area.

According to the illustrated embodiment of FIG. 2 , the dummy pads 415 b are isolated, inactive pads, which are not electrically connected to other circuitry in the RDL 410 . In other words, no signal will pass through these dummy pads 415 b during operation of the chip package.

As shown in FIG. 3 , a plurality of bumps 416 a and 416 b , such as micro-bumps, may be formed on the RDL 410 for further connections. The bumps 416 a may be directly formed on respective bump pads 415 a in the metal layer 414 . The bumps 416 b may be directly formed on respective dummy pads 415 b in the metal layer 414 . In some embodiments, a passivation layer or a dielectric layer (not shown) may be formed on the RDL 410 before the formation of the bumps 416 a and 416 b.

As shown in FIG. 4 , after the formation of the bumps 416 a and 416 b , individual active chips 420 a (e.g., flip-chips or dies) with their active sides facing down toward the RDL 410 are then mounted on the RDL 410 through the bumps 416 a to thereby forming a stacked chip-to-wafer (C2 W) construction. These individual flip-chips or dies 420 a are active integrated circuit chips with certain functions, for example, GPU (graphics processing unit), CPU (central processing unit), memory chips, etc.

According to the illustrated embodiment, the dummy chips 420 b are mounted in a peripheral area 104 around a chip mounting area 102 through the bumps 416 b . FIG. 8 illustrates an example of the arrangement of the active chips 420 a and the dummy chips 420 b . For example, the dummy chips 420 b may be dummy silicon chips, dies or pieces having dimensions or sizes similar to that of the active chips 420 a , but not limited thereto. It is to be understood that other materials such as metal, glass or ceramic may be used.

Optionally, an underfill (not shown) may be applied under each of the active chips 420 a and the dummy chips 420 b . Thereafter, a thermal process may be performed to reflow the bumps 416 a and 416 b.

As shown in FIG. 5 , after the die-bonding process, a molding compound 500 is applied. The molding compound 500 covers the attached active chips 420 a and the dummy chips 420 b and the top surface of the RDL 410 . The molding compound 500 may be subjected to a curing process. The molding compound 500 may comprise a mixture of epoxy and silica fillers, but is not limited thereto.

Optionally, a top portion of the molding compound 500 may be polished away to expose a top surfaces of the active chips 420 a and the dummy chips 420 b.

Since most of the peripheral area around the chip mounting area is occupied by the dummy chips 420 b , the used amount of the molding compound 500 is reduced, and therefore the warpage of the substrate or wafer is alleviated or avoided. According to the illustrated embodiment, these dummy chips 420 b may also be referred to as “warpage-control” dummy chips.

As shown in FIG. 6 , after the formation of the molding compound 500 , the carrier 300 is removed or peeled off to expose the passivation layer 310 , thereby forming a TSV-less interposer 301 . The de-bonding of the carrier 300 may be performed by using a laser process or UV irradiation process, but not limited thereto.

To peel off the carrier 300 , another temporary carrier substrate (not shown) may be attached to the molding compound 500 . After the de-bonding of the carrier 300 , openings may be formed in the passivation layer 310 to expose respective solder pads, and then solder bumps or solder balls 520 maybe formed on the respective solder pads.

Thereafter, as shown in FIG. 7 , a dicing process is performed to separate individual wafer-level packages 10 from one another.

With reference to FIGS. 9 through 13 , depicted are schematic diagrams showing an exemplary method for fabricating a wafer-level package (WLP) with a substrate-less (or TSV-less) interposer according to another embodiment of the invention, wherein like numeral numbers designate like regions, layers or elements.

As shown in FIG. 9 , likewise, a redistribution layer (RDL) 410 is formed on the passivation layer 310 . The RDL 410 may comprise at least one dielectric layer 412 and at least one metal layer 414 . The dielectric layer 412 may comprise organic materials such as polyimide (PI) or inorganic materials such as silicon nitride, silicon oxide, or the like, but not limited thereto. The metal layer 414 may comprise aluminum, copper, tungsten, titanium, titanium nitride, or the like.

According to the illustrated embodiment of FIG. 9 , the metal layer 414 may comprise a plurality of bump pads 415 exposed from a top surface of the dielectric layer 412 . The bump pads 415 are disposed within a chip mounting area 102 . In general, no bump pads are formed within a peripheral area 104 around the chip mounting area 102 . A plurality of bumps 416 , such as micro-bumps, may be formed on the RDL 410 for further connections. The bumps 416 may be directly formed on respective bump pads 415 in the metal layer 414 .

As shown in FIG. 10 , after the formation of the bumps 416 , individual flip-chips or dies 420 a with their active sides facing down toward the RDL 410 are then mounted on the RDL 410 through the bumps 416 , thereby forming a stacked chip-to-wafer (C2 W) construction. Optionally, an underfill (not shown) may be applied under each active chip 420 a . Thereafter, a thermal process may be performed to reflow the bumps 416 .

According to the illustrated embodiment, dummy chips 420 b are mounted on the dielectric layer 412 of the RDL 410 within the peripheral area 104 around the chip mounting area 102 by using an adhesive 430 .

As shown in FIG. 11 , after the die-bonding process, a molding compound 500 is applied. The molding compound 500 covers the attached active chips 420 a and the dummy chips 420 b and the exposed top surface of the RDL 410 . The molding compound 500 may be subjected to a curing process. Optionally, a top portion of the molding compound 500 may be polished away to expose top surfaces of the active chips 420 a and the dummy chips 420 b.

As shown in FIG. 12 , after the formation of the molding compound 500 , the carrier 300 is removed or peeled off to expose the passivation layer 310 , thereby forming a TSV-less interposer 301 . The de-bonding of the carrier 300 may be performed by using a laser process or UV irradiation process, but is not limited thereto.

To peel off the carrier 300 , another temporary carrier substrate (not shown) may be attached to the molding compound 500 . After the de-bonding of the carrier 300 , openings may be formed in the passivation layer 310 to expose respective solder pads, and then solder bumps or solder balls 520 maybe formed on the respective solder pads. The temporary carrier substrate is then removed.

Thereafter, as shown in FIG. 13 , a dicing process is performed to separate individual wafer-level packages 10 a from one another.

Those skilled in the art will readily observe that numerous modifications and alterations of the devices and methods may be made while retaining the teachings of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.