Computing architectures continue to scale to smaller form factors while pushing towards higher bandwidths and computing capacity. One solution for enabling such design goals is to use chiplet architectures. Instead of a single large chip, a plurality of smaller chiplets are stitched together by a bridge. When the bridge is embedded in the underlying package substrate, the bridge may be referred to as an embedded bridge solution. Existing bridge solutions typically do not allow for power to pass through a thickness of the bridge. Instead, traces are routed over the bridge in order to provide power within the footprint of the bridge. This complicates routing and increases the length of the power delivery path, which can impact performance.
Accordingly, some solutions have proposed using vias through a thickness of the bridge. This allows for power to pass directly through the bridge and reduces path length and mitigates routing complexity. However, the integration of such bridges into the package substrate is not without issue. One issue that arises with such architectures is that the current capacity passing through the bridge is limited by traditional solder interconnect solutions. Additionally, the solder increases the standoff height of the bridge. This increases the depth of the cavity in the package substrate that is used to accommodate the bridge. Another issue that may arise is the underfill uniformity. Particularly, tight pitches and small gaps makes it difficult to dispense underfill material between the bridge and the bottom of the cavity. This leads to voids which can pose reliability issues for the package substrate.
Described herein are electronic systems, and more particularly, architectures for coupling an embedded bridge with vias to an underlying package substrate, in accordance with various embodiments. In the following description, various aspects of the illustrative implementations will be described using terms commonly employed by those skilled in the art to convey the substance of their work to others skilled in the art. However, it will be apparent to those skilled in the art that the present disclosure may be practiced with only some of the described aspects. For purposes of explanation, specific numbers, materials and configurations are set forth in order to provide a thorough understanding of the illustrative implementations. However, it will be apparent to one skilled in the art that the present disclosure may be practiced without the specific details. In other instances, well-known features are omitted or simplified in order not to obscure the illustrative implementations.
Various operations will be described as multiple discrete operations, in turn, in a manner that is most helpful in understanding the present disclosure, however, the order of description should not be construed to imply that these operations are necessarily order dependent. In particular, these operations need not be performed in the order of presentation.
Various embodiments or aspects of the disclosure are described herein. In some implementations, the different embodiments are practiced separately. However, embodiments are not limited to embodiments being practiced in isolation. For example, two or more different embodiments can be combined together in order to be practiced as a single device, process, structure, or the like. The entirety of various embodiments can be combined together in some instances. In other instances, portions of a first embodiment can be combined with portions of one or more different embodiments. For example, a portion of a first embodiment can be combined with a portion of a second embodiment, or a portion of a first embodiment can be combined with a portion of a second embodiment and a portion of a third embodiment.
As noted above, embedded bridge architectures have been used in order to implement device scaling which can lead to smaller devices while maintaining or improving device performance. However, continued scaling of embedded bridge structures has led to greater issues with electrically coupling the bridge to other features within the package substrate.
An example of a typical embedded bridge structure is shown in
In an embodiment, the bridge 120 is embedded within the buildup layers 114. The bridge 120 may also be referred to as a “die” or a “bridge die” in some embodiments. The bridge 120 may be a dimensionally stable material. For example, the bridge 120 may comprise silicon, other semiconductor materials, a ceramic, glass, or the like. In an embodiment, electrically conductive routing (e.g., traces, pads, etc.) may be provided on the bridge 120. For example, pads 123 are shown in
In the illustrated embodiment, there are no vias through a thickness of the bridge 120. Accordingly, power is not able to be routed through the bridge 120. Instead, power is provided in a path that passes adjacent to a sidewall of the bridge 120. Once above the level of the top surface of the bridge 120, a trace 117 can route power into the footprint of the bridge 120. This increases the length of the power delivery path and decreases performance. Additionally, the lateral routing makes routing within the package substrate 110 more complicated.
Accordingly, embodiments disclosed herein may utilize a bridge 120 that includes vias 124. An example of such an embodiment is shown in
When a glass core 112 is used, thickness variation is typically improved. The improvement is maximized as the bridge 120 is moved closer to the surface of the glass core 112. That is, reducing the thickness of the buildup layers 114 between the bridge 120 and the core 112 is beneficial. However, as the bridge 120 is moved closer to the core 112, potential for damaging the core 112 is increased. The core 112 is brittle and is prone to cracking or other damage. Accordingly, some amount of buffer layer is currently necessary between the core 112 and the bottom of the bridge 120.
However, the solder 126 between the bridge 120 and the pad 115 on the buildup layers 114 may result in several issues. For example, the solder 126 has poor current carrying capability, which leads to a less than desirable IMAX value. Additionally, underfill may be needed around the solder 126. Due to tight pitch between interconnects, voids may be formed. Voids negatively impact product reliability. Solder 126 also increases the standoff height of the device, and the cavity to accommodate the bridge 120 needs to be deeper.
Therefore, embodiments disclosed herein provide improved interconnect architectures in order to address these issues. In one embodiment, a hybrid bonding architecture is used to couple the bridge to the bottom surface of the cavity in the buildup layers. A hybrid bonding architecture may use direct copper-to-copper bonding and dielectric-to-dielectric (or silicon-to-dielectric) bonding around the pads. In an embodiment, the hybrid bonding may incorporate dielectric material modification. For example, silica spheres may be embedded in the dielectric layer to enable improved compressibility. A vapor infiltration process may also be used to integrate inorganic particles into the dielectric layer in order to modify coefficient of thermal expansion (CTE). A B-stage dielectric may also be used in some embodiments. In an additional embodiment, the pads of the hybrid bonding architecture may include gold and/or silver (either as the entire pad or as a liner over the pad).
In other instances, a porous bump architecture may be used instead of hybrid bonding. The porous bump architecture may include metals that have a higher electrical conductivity than solder. Also, the porous bump architecture may allow for a greater degree of freedom for placement of the bridge in the cavity. As such, assembly processes are simplified.
Referring now to
Referring now to
The core 212 may be substantially all glass. The core 212 may be a solid mass comprising a glass material with an amorphous crystal structure where the solid glass core may also include various structures—such as vias, cavities, channels, or other features—that are filled with one or more other materials (e.g., metals, metal alloys, dielectric materials, etc.). As such, core 212 may be distinguished from, for example, the “prepreg” or “RF4” core of a Printed Circuit Board (PCB) substrate which typically comprises glass fibers embedded in a resinous organic material, such as an epoxy.
The core 212 may have any suitable dimensions. In a particular embodiment, the core 212 may have a thickness that is approximately 50 μm or greater. For example, the thickness of the core 212 may be between approximately 50 μm and approximately 1.4 mm. Though, smaller or larger thicknesses may also be used. The core 212 may have edge dimensions (e.g., length, width, etc.) that are approximately 10 mm or greater. For example, edge dimensions may be between approximately 10 mm to approximately 250 mm. Though, larger or smaller edge dimensions may also be used. More generally, the area dimensions of the core 212 (from an overhead plan view) may be between approximately 10 mm×10 mm and approximately 250 mm×250 mm. In an embodiment, the core 212 may have a first side that is perpendicular or orthogonal to a second side. In a more general embodiment, the core 212 may comprise a rectangular prism volume with sections (e.g., vias) removed and filled with other materials (e.g., metal, etc.).
The core 212 may comprise a single monolithic layer of glass. In other embodiments, the core 212 may comprise two or more discrete layers of glass that are stacked over each other. The discrete layers of glass may be provided in direct contact with each other, or the discrete layers of glass may be mechanically coupled to each other by an adhesive or the like. The discrete layers of glass in the core 212 may each have a thickness less than approximately 50 μm. For example, discrete layers of glass in the core 212 may have thicknesses between approximately 25 μm and approximately 50 μm. Though, discrete layers of glass may have larger or smaller thicknesses in some embodiments. As used herein, “approximately” may refer to a range of values within ten percent of the stated value. For example approximately 50 μm may refer to a range between 45 μm and 55 μm.
The core 212 may be any suitable glass formulation that has the necessary mechanical robustness and compatibility with semiconductor packaging manufacturing and assembly processes. For example, the core 212 may comprise aluminosilicate glass, borosilicate glass, alumino-borosilicate glass, silica, fused silica, or the like. In some embodiments, the core 212 may include one or more additives, such as, but not limited to, Al2O3, B2O3, MgO, CaO, SrO, BaO, SnO2, Na2O, K2O, SrO, P2O3, ZrO2, Li2O, Ti, or Zn. More generally, the core 212 may comprise silicon and oxygen, as well as any one or more of aluminum, boron, magnesium, calcium, barium, tin, sodium, potassium, strontium, phosphorus, zirconium, lithium, titanium, or zinc. In an embodiment, the core 212 may comprise at least 23 percent silicon (by weight) and at least 26 percent oxygen (by weight). In some embodiments, the core 212 may further comprise at least 5 percent aluminum (by weight).
In an embodiment, vias 205 may pass through a thickness of the core 212. The vias 205 may comprise vertical sidewalls, tapered sidewalls, or any other suitable cross-sectional shape. In the particular embodiment shown in
In an embodiment, the package substrate 210 may comprise a cavity 230 that is formed into the buildup layers 214. The cavity 230 may be formed with an etching process, a laser ablation process, or the like. In the illustrated embodiment, the cavity 230 has vertical sidewalls. In other instances, the cavity 230 may have sloped sidewalls. The cavity 230 may pass partially through a thickness of the buildup layers 214. That is, a portion of the buildup layers 214 remains between the bottom surface of the cavity 230 and the top surface of the core 212. In other embodiments, the cavity 230 may pass entirely through a thickness of the buildup layers 214.
Pads 215 may be exposed at a bottom of the cavity 230. The pads 215 may extend up from the bottom surface of the cavity 230, as shown in
In an embodiment, first portions 226A of an interconnect may be provided on the pads 215 in the cavity 230. The first portions 226A may include an electrically conductive material with a first composition and a first structure. For example, the first composition may include one or more electrically conductive metals, such as, but not limited to, tin, copper, nickel, silver, or gold. The first structure may include a porous structure. As used herein, a porous structure refers to a material that has a matrix material (e.g., a metal) and pores (or voids) distributed through the matrix material. The pores or voids may be filled with air or another gas/fluid. In an embodiment, the porous structure may have a porosity that is up to approximately 50% porosity. That is, when viewing a cross-section of the porous first portions 226A, an area of the voids may account for up to approximately 50% of the total area of the first portions 226A.
Referring now to
In an embodiment, pads 225 may be provided at the bottom of the bridge 220. The pads 225 may extend out from the bottom surface of the bridge 220. In other instances, the pads 225 may be flush with the bottom surface of the bridge 220. That is, bottom surfaces of the pads 225 may be substantially coplanar with the bottom surface of the bridge 220. In an embodiment, a second portion 226B of the interconnect may be provided on the pads 225.
The second portion 226B of the interconnect may have a second material composition and a second structure. The second material composition may be different than the first material composition of the first portion 226A. For example, the second material composition may include a solder (e.g., a tin-based solder) or the like. Additionally, the second structure may be different than the first structure. For example, the second structure may be substantially non-porous in some embodiments. Other embodiments may include a second portion 226B with a porosity up to approximately 30%.
Referring now to
In an embodiment, the remainder of the cavity 230 is unfilled. In other embodiments, an underfill or the like may be provided between the interconnects and/or around the bridge 220 in the cavity 230. The top of the bridge 220 may be substantially coplanar with the top of the buildup layers 214. Though, the top of the bridge 220 may be above the buildup layers 214, or the top of the bridge 220 may be below the top of the buildup layers 214.
Referring now to
Referring now to
The first composition may be different than the second composition. The first composition of the first portion 326A may comprise one or more metals, such as, but not limited to, tin, copper, nickel, silver, or gold. The second composition of the second portion 326B may be a standard solder based material, such as a tin-based solder or the like. In an embodiment, the second structure is different than the first structure. For example, a porosity of the first structure is higher than a porosity of the second structure. In an embodiment, the porosity of the first portion 326A may be up to approximately 50% in some embodiments. As shown, the first portion 326A may include a metal matrix 303 with a plurality of voids 305 distributed through the matrix 303. The voids 305 may be air filled, or filled with any gas/fluid material.
As shown in
Referring now to
Referring now to
Referring now to
Referring now to
Referring now to
Referring now to
Referring now to
Referring now to
Referring now to
Referring now to
The pads 425 may be in direct contact with the pads 415. That is, there may not be any solder or the like between the pads 415 and pads 425. Similarly, the bottom surface of the bridge 420 may be in direct contact with the buildup layers 414 at the bottom surface of the cavity 430. That is, a hybrid bonding architecture may be provided that includes copper-to-copper bonds (i.e., pads 415 to pads 425) and dielectric-to-bridge bonds (i.e., buildup layers 414 to bridge 420). The material of the bridge 420 that contacts the bottom surface of the cavity 430 may be silicon or an intervening dielectric layer (not shown). In some embodiments, a treatment of the bottom surface of the cavity 430 and/or the pads 415 may occur before attaching the bridge 420. For example, a plasma treatment process may be used in some embodiments.
Referring now to
In an embodiment, the package substrate 510 comprises a core 512 and buildup layers 514 over the core 512. In an embodiment, the core 512 is a glass core or an organic core. A glass core 512 may be similar to any of the core architectures described in greater detail herein. Vias 505 may pass through a thickness of the core 512. The vias 505 may have any suitable cross-sectional shape, such as those described in greater detail herein. Electrically conductive routing (e.g., vias 513, pads 515, and traces) may be provided on and/or in the buildup layers 514.
In an embodiment, a bridge 520 may be at least partially embedded in the buildup layers 514 over the core 512. The bridge 520 may comprise vias 524 (e.g., TSVs). Electrically conductive routing, such as traces 529 and the like, may couple pads 523 together on the top of the bridge 520. In this way, overlying dies (not shown) can be communicatively coupled to each other through the bridge 520.
In an embodiment, the bridge 520 may be coupled to the buildup layers 514 through a bump-to-bump, a bump-to-pad, or a pad-to-pad interface (which may sometimes be referred to as a hybrid bonding interface or architecture). The bump 525 on the bridge 520 side may directly contact a pad 515 on the buildup layers 514. As shown, the shading of bumps 525 and pads 515 are different than the remainder of the electrical routing in the buildup layers 514 in order to indicate a different material is used. For example, the bumps 525 and the pads 515 may comprise one or more high current carrying capacity materials, such as, but not limited to, gold or silver.
In an embodiment, the bumps 525 are narrower than the pads 515. Though, in other instances, the bumps 525 and the pads 515 may have substantially similar widths, or the bump 525 may be wider than the pads 515. The bridge 520 may be in direct contact with the buildup layers 514. Other embodiments may include a dielectric layer (not shown) between the bridge 520 and the buildup layers 514.
Referring now to
In an embodiment, the bump 525 and the pad 515 may comprise the same or similar material as the remainder of the electrical routing (not shown in
Referring now to
In an embodiment, a bridge 620 is inserted into a cavity 630 into the buildup layers 614. The bridge 620 may be similar to any bridge architecture described in greater detail herein. In an embodiment, the bridge 620 comprises vias 624 (e.g., TSVs) with pads 625 at a bottom of the bridge 620. In an embodiment, the pads 625 may be bonded to pads 615 by an interconnect that includes a first bump 651 and a second bump 652. The bumps 651 and 652 may comprise copper or the like. The bumps 651 and 652 may be surrounded by a dielectric layer 609. Between the dielectric layer 609 and the bumps 651 and 652, the interconnect architecture between the bridge 620 and the buildup layer 614 may be considered a hybrid bonding architecture.
Referring now to
Referring now to
The filler particles 755 allow for improved compression of the dielectric layer 709 during bonding operations. For example, the filler particles 755 may be hollow spheres (e.g., air filled spheres) that are deformable or prone to fracture when exposed to compressive forces. The shell of the spheres may comprise silica or the like. The filler particles 755 may occupy up to approximately 50% of an area of the dielectric layer 709. A patterning layer 753 may be provided over the dielectric layer 709.
Referring now to
Referring now to
During the bonding process (e.g., a thermocompression bonding (TCB) process), the dielectric layers 709 may be compressed to allow first bumps 751 to directly contact second bumps 752. The compression may be enabled through the deformation of the filler particles 755. For example, the filler particles 755 may be deformed, cracked, or otherwise damaged.
Referring now to
Referring now to
Referring now to
Referring now to
The filler particles 857 may be inorganic fillers that are generated from metalorganic precursor materials in the vapor infiltration process 856. The inclusion of filler particles 857 may provide several benefits. For one, the mechanical robustness of the dielectric layer 809 is increased. Additionally, the filler particles 857 may be used to modulate the CTE of the dielectric layer 809 to more closely match that of the first bumps 851.
Referring now to
During the bonding process (e.g., a TCB process), the opposing dielectric layers 809 may be bonded together. Similarly, the first bumps 851 may be bonded to the second bumps 852. As such, a dielectric-to-dielectric bond and a metal-to-metal bond is provided to generate a hybrid bonding interface.
Referring now to
Referring now to
Referring now to
Referring now to
Referring now to
Referring now to
It is to be appreciated that the three embodiments in
Referring now to
In an embodiment, the package substrate 1010 may be similar to any of the package substrates described in greater detail herein. For example, the package substrate 1010 may comprise a core 1012 with buildup layers 1014 above and below the core 1012. The core 1012 may be similar to any of the cores described in greater detail herein. In an embodiment, a bridge 1020 is embedded in a cavity 1030 into the buildup layers 1014. The bridge 1020 may include vias 1024 (e.g., TSVs) that end at pads 1025. The pads 1025 may be directly bonded to pads 1015 on the bottom surface of the cavity 1030. A dielectric layer 1009 or underfill may surround the interconnects below the bridge 1020. The interconnect between the bridge 1020 and the buildup layers 1014 may be referred to as a hybrid bonding interconnect in some embodiments. Though, solder (e.g., a single solder, or a solder and a porous bump) may also be used as the interconnect in other embodiments. A solder resist 1035 may be provided at the top of the package substrate 1010.
In an embodiment, a pair of dies 1095 may be communicatively coupled together through the bridge 1020. For example, first level interconnects (FLIs) 1094 may couple the dies 1095 to the package substrate 1010 and the bridge 1020. The dies 1095 may comprise a central processing unit (CPU), a graphics processing unit (GPU), an XPU, a communications die, a memory die, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or the like.
These other components include, but are not limited to, volatile memory (e.g., DRAM), non-volatile memory (e.g., ROM), flash memory, a graphics processor, a digital signal processor, a crypto processor, a chipset, an antenna, a display, a touchscreen display, a touchscreen controller, a battery, an audio codec, a video codec, a power amplifier, a global positioning system (GPS) device, a compass, an accelerometer, a gyroscope, a speaker, a camera, and a mass storage device (such as hard disk drive, compact disk (CD), digital versatile disk (DVD), and so forth).
The communication chip 1106 enables wireless communications for the transfer of data to and from the computing device 1100. The term “wireless” and its derivatives may be used to describe circuits, devices, systems, methods, techniques, communications channels, etc., that may communicate data through the use of modulated electromagnetic radiation through a non-solid medium. The term does not imply that the associated devices do not contain any wires, although in some embodiments they might not. The communication chip 1106 may implement any of a number of wireless standards or protocols, including but not limited to Wi-Fi (IEEE 802.11 family), WiMAX (IEEE 802.16 family), IEEE 802.20, long term evolution (LTE), Ev-DO, HSPA+, HSDPA+, HSUPA+, EDGE, GSM, GPRS, CDMA, TDMA, DECT, Bluetooth, derivatives thereof, as well as any other wireless protocols that are designated as 3G, 4G, 5G, and beyond. The computing device 1100 may include a plurality of communication chips 1106. For instance, a first communication chip 1106 may be dedicated to shorter range wireless communications such as Wi-Fi and Bluetooth and a second communication chip 1106 may be dedicated to longer range wireless communications such as GPS, EDGE, GPRS, CDMA, WiMAX, LTE, Ev-DO, and others.
The processor 1104 of the computing device 1100 includes an integrated circuit die packaged within the processor 1104. In some implementations of the disclosure, the integrated circuit die of the processor may be part of an electronic package that includes a bridge that is bonded to the package substrate through a hybrid bonding interface, in accordance with embodiments described herein. The term “processor” may refer to any device or portion of a device that processes electronic data from registers and/or memory to transform that electronic data into other electronic data that may be stored in registers and/or memory.
The communication chip 1106 also includes an integrated circuit die packaged within the communication chip 1106. In accordance with another implementation of the disclosure, the integrated circuit die of the communication chip may be part of an electronic package that includes a bridge that is bonded to the package substrate through a hybrid bonding interface, in accordance with embodiments described herein.
In an embodiment, the computing device 1100 may be part of any apparatus. For example, the computing device may be part of a personal computer, a server, a mobile device, a tablet, an automobile, or the like. That is, the computing device 1100 is not limited to being used for any particular type of system, and the computing device 1100 may be included in any apparatus that may benefit from computing functionality.
The above description of illustrated implementations of the disclosure, including what is described in the Abstract, is not intended to be exhaustive or to limit the disclosure to the precise forms disclosed. While specific implementations of, and examples for, the disclosure are described herein for illustrative purposes, various equivalent modifications are possible within the scope of the disclosure, as those skilled in the relevant art will recognize.
These modifications may be made to the disclosure in light of the above detailed description. The terms used in the following claims should not be construed to limit the disclosure to the specific implementations disclosed in the specification and the claims. Rather, the scope of the disclosure is to be determined entirely by the following claims, which are to be construed in accordance with established doctrines of claim interpretation.
Example 1: an apparatus, comprising: a substrate, wherein the substrate comprises a dielectric material; a cavity into a surface of the substrate; a first pad on a bottom surface of the cavity; a die at least partially in the cavity; a via that passes through at least a portion of a thickness of the die; and a second pad on the die, wherein the second pad directly contacts the first pad, and wherein the first pad is the only electrically conductive structure between the via and the second pad.
Example 2: the apparatus of Example 1, wherein a surface of the die directly contacts the bottom surface of the cavity.
Example 3: the apparatus of Example 1 or Example 2, wherein one or both of the first pad and the second pad comprise gold or silver.
Example 4: the apparatus of Examples 1-3, further comprising: a layer between the die and the bottom surface of the cavity.
Example 5: the apparatus of Example 4, wherein a spheroidal shell is embedded in the layer.
Example 6: the apparatus of Example 5, wherein the spheroidal shell is cracked.
Example 7: the apparatus of Examples 4-6, further comprising inorganic fillers embedded in the layer.
Example 8: the apparatus of Examples 4-7, wherein the layer is a B-stage material.
Example 9: the apparatus of Examples 1-8, wherein the substrate is over a core.
Example 10: the apparatus of claim 9, wherein the core comprises a solid glass layer with a rectangular prism volume.
Example 11: an apparatus, comprising: a substrate; a cavity into a surface of the substrate; a first pad on a bottom surface of the cavity; a die at least partially within the cavity; a second pad on the die; and an interconnect between the first pad and the second pad, wherein the interconnect comprises a first portion with a first material composition and a second portion with a second material composition.
Example 12: the apparatus of Example 11, wherein the first portion comprises a porous metal.
Example 13: the apparatus of Example 12, wherein the porous metal comprises one or more of tin, copper, or nickel.
Example 14: the apparatus of Example 12 or Example 13, wherein up to approximately 50% of a cross-sectional area of the porous metal comprises air.
Example 15: the apparatus of Examples 11-14, wherein an interface between the first portion and the second portion is curved.
Example 16: the apparatus of Examples 11-15, wherein the first portion comprises a lateral protrusion.
Example 17: an apparatus, comprising: a board; a package substrate over the board, wherein the package substrate comprises: a core; buildup layers over the core; and a bridge with a via embedded in the buildup layers, wherein the bridge is bonded to the buildup layers with a hybrid bonding architecture; a first die over the package substrate; and a second die over the package substrate, wherein the first die is communicatively coupled to the second die by the bridge.
Example 18: the apparatus of Example 17, wherein the core comprises glass with a rectangular prism shape.
Example 19: the apparatus of Example 17 or Example 18, wherein the hybrid bonding architecture comprises a copper-to-copper bond and a dielectric-to-dielectric bond.
Example 20: the apparatus of Examples 17-19, wherein the apparatus is part of a personal computer, a server, a mobile device, a tablet, or an automobile.