This application claims the benefit under 35 USC 119 (a) of Korean Patent Application No. 10-2023-0144560, filed on Oct. 26, 2023, in the Korean Intellectual Property Office, the entire disclosure of which is incorporated herein by reference for all purposes.
Embodiments of the present disclosure relate to a semiconductor package and a method of manufacturing a semiconductor package.
According to the trend for miniaturization and higher performance of semiconductor packages, the development of system-in-package (SiP) technology in which a plurality of semiconductor chips performing different functions are embedded in a single package is required. To form fine wirings connecting semiconductor chips within a package, a technology is being used to form a Through Silicon Via (TSV) and bond semiconductor chips to each other through bonding pads.
Example embodiments of the present disclosure provide a semiconductor package and a method of manufacturing a semiconductor package with improved electrical characteristics and/or reliability of a front surface of a semiconductor chip.
According to embodiments of the present discourse, a semiconductor package is provided and includes a plurality of semiconductor chips bonded to each other through direct bonding, the plurality of semiconductor chips including a first semiconductor chip and a second semiconductor chip. The first semiconductor chip including: a front insulating layer bonded to a back insulating layer of the second semiconductor chip; a front pad surrounded by the front insulating layer; a device layer on a back surface of the front insulating layer and including an interconnection structure electrically connected to the front pad; a conductive pattern between the interconnection structure and the front pad; and a support insulating layer between the device layer and the front insulating layer and surrounding the conductive pattern, wherein a gap is between a first side surface of the support insulating layer and a second side surface of the conductive pattern that faces the first side surface.
According to embodiments of the present discourse, a semiconductor package is provided and includes a semiconductor chip including: a front insulating layer; a front pad surrounded by the front insulating layer; a device layer on a back surface of the front insulating layer and including an interconnection structure electrically connected to the front pad; a conductive pattern between the interconnection structure and the front pad; a support insulating layer between the device layer and the front insulating layer and surrounding the conductive pattern; a semiconductor substrate on a back surface of the device layer; a through-electrode penetrating through the semiconductor substrate; and a back insulating layer on a back surface of the semiconductor substrate, wherein at least one from among a first side surface of the support insulating layer and a second side surface of the conductive pattern that faces the first side surface is curved more than each of a third side surface of the front insulating layer and a fourth side surface of the front pad that faces the third side surface.
According to embodiments of the present disclosure, a method of manufacturing a semiconductor package and includes: forming a conductive pattern on a front surface of a device layer of a first semiconductor chip; disposing an insulating material on the front surface of the device layer; annealing and cooling a combination portion of the conductive pattern and the insulating material; forming a support insulating layer by flattening the insulating material; forming a front pad and a front insulating layer on a front surface of the conductive pattern and a front surface of the support insulating layer; and direct bonding the front insulating layer of the first semiconductor chip to a back insulating layer of a second semiconductor chip.
The above and other aspects, features, and advantages of example embodiments of the present disclosure will be more clearly understood from the following detailed description, taken in conjunction with the accompanying drawings, in which:
The below description refers to the accompanying drawings, which illustrate, by way of example, non-limiting example embodiments of the present disclosure. These example embodiments are described in sufficient detail to enable those skilled in the art to practice embodiments of the present disclosure. It should be understood that the various embodiments are different from one another but are not necessarily mutually exclusive. For example, specific shapes, structures and characteristics described herein with respect to one embodiment may be implemented in other embodiments without departing from the spirit and scope of the present disclosure. Additionally, it should be understood that the location or arrangement of individual components within each described embodiment may be changed without departing from the spirit and scope of the present disclosure. Accordingly, the detailed description below is not intended to be taken in a limiting sense. Similar reference numbers in the drawings refer to identical or similar components and/or functions across various aspects.
It will be understood that when an element is referred to as being “on,” “connected to,” or “coupled to” another element, it can be directly on, connected to, or coupled to the other element or intervening elements may be present. In contrast, when an element is referred to as being “directly on,” “directly connected to,” or “directly coupled to” another element, there are no intervening elements present.
Referring to
The plurality of semiconductor chips (e.g., semiconductor chips 100A1, 100A2, 100A3, and 100C) may be memory chips. For example, the memory chip may be a volatile memory chip such as a Dynamic Random Access Memory (DRAM) or a Static Random Access Memory (SRAM), or may be a non-volatile memory chip such as a Phase-change Random Access Memory (PRAM), a Magnetoresistive Random Access Memory (MRAM), a Ferroelectric Random Access Memory (FeRAM) or a Resistive Random Access Memory (RRAM). Alternatively, some of the plurality of semiconductor chips (e.g., semiconductor chips 100A1, 100A2, 100A3, and 100C) may be memory chips and others may be logic chips. The logic chip may be, for example, a microprocessor, analog element, or digital signal processor, and may control the operation of memory chips. For example, a combination of the plurality of semiconductor chips (e.g., semiconductor chips 100A1, 100A2, 100A3, and 100C) may be a High Bandwidth Memory (HBM) DRAM.
The semiconductor substrate 110 and the semiconductor substrate 110′ may include a semiconductor such as silicon. For example, the semiconductor substrate 110 may include various impurity regions for individual devices and a device isolation structure such as a shallow trench isolation (STI) structure. The semiconductor is not limited to silicon and may include at least one from among germanium, silicon carbide (SiC), gallium arsenide (GaAs), indium arsenide (InAs), and indium phosphide (InP). For example, the device may include a Planar Metal Oxide Semiconductor FET (MOSFET), a FinFET of which an active region has a fin structure, a Multi Bridge Channel FET (MBCFET™) or a Gate-All-Around transistor including a plurality of channels stacked vertically on the active region, or a Vertical FET (VFET), but embodiments of the present disclosure are not limited thereto.
Each of the through-electrodes 130 may have a pillar structure penetrating through the semiconductor substrate 110. The through-electrodes 130 may not penetrate through the semiconductor substrate 110′. The upper end of the through-electrode 130 may be connected to the back pad 154, and the lower end may be electrically connected to the front pad 152 through an interconnection structure 140. The through-electrode 130 may include a via plug 135 and an insulating liner 131 surrounding the via plug 135. The insulating liner 131 may electrically separate the via plug 135 from the semiconductor substrate 110.
The device layer 120 may include an interconnection structure 140 connected to a plurality of individual devices formed on the front (lower) surface of the semiconductor substrate 110 and the semiconductor substrate 110′. The interconnection structure 140 may include an interconnection layer 142 and an interconnection via 145. For example, the interconnection structure 140 may have a structure in which a plurality of interconnection layers 142 are stacked in the Z direction, and may include a plurality of interconnection vias 145 that electrically connect the plurality of interconnection layers 142 in the Z direction. The interconnection structure 140 may be electrically connected to the front pad 152 disposed below the device layer 120. The interconnection layer 142 and the interconnection via 145 may include at least one from among copper (Cu), copper alloy, aluminum (Al), and aluminum alloy. The metal material is not limited thereto, and may be implemented as at least one from among nickel (Ni), gold (Au), cobalt (Co), tantalum (Ta), tellurium (Te), titanium (Ti), tungsten (W) or alloys thereof (for example, TiN, TaN). A space in the device layer 120 where the interconnection structure 140 is not disposed may be filled with an insulating layer. For example, the insulating layer may include at least one from among silicon oxide (SiO), silicon nitride (SiN), silicon oxynitride (SiON), silicon oxycarbide (SiOC), and silicon carbonitride (SiCN).
The front pad 152 may be arranged on the front (lower) surface of each of the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C), and electrical paths to the outside of each of the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C) may be provided. The exterior of each of the semiconductor chips 100A2, 100A3, and 100C may be the back pad 154 of the semiconductor chips 100A1, 100A2, and 100A3 immediately below. For example, the front pad 152 is connected to the back pad 154, thereby providing an electrical connection path between the semiconductor chips 100A1, 100A2, 100A3, and 100C. The back pad 154 may be arranged on the back (top) of each of the semiconductor chips 100A1, 100A2, and 100A3, and may be connected between the front pad 152 and the through-electrodes 130.
The front (lower surface) and back surface (upper surface) of each of the front pad 152 and the back pad 154 may be polygonal or circular, and the interconnection layer 142 may have a wider width than the line width of the interconnection. Each of the front pad 152 and the back pad 154 may include a metal material that has high conductivity and may be bonded to each other, such as copper (Cu) or a copper alloy. The metal material is not limited to copper, and may be implemented as at least one from among aluminum (Al), nickel (Ni), gold (Au), cobalt (Co), tantalum (Ta), tellurium (Te), titanium (Ti), tungsten (W) or alloys thereof. For example, the front pad 152 and the back pad 154 may be temporarily bonded to make direct contact and then firmly bonded by mutual diffusion of copper through a high-temperature annealing process.
The front insulating layer 162b may surround the front pad 152, and the back insulating layer 164 may surround the back pad 154. The front surfaces (lower surfaces) and back surfaces (upper surfaces) of the front insulating layer 162b and the front pad 152 may each be coplanar, and the front surfaces (lower surfaces) and back surfaces (upper surfaces) of the back insulating layer 164 (or a second insulating layer 164b thereof) and the back pad 154 may each be coplanar.
Each of the front insulating layer 162b and the back insulating layer 164 may include SiO2. When the front insulating layer 162b and the back insulating layer 164 are bonded to each other, the oxygen of SiO2 may form a covalent bond to silicon. Accordingly, the front insulating layer 162b and the back insulating layer 164 may have strong bonding strength. And the insulating material forming the front insulating layer 162b and the back insulating layer 164 is not limited to silicon oxide, and may be implemented with at least one from among SiN, SiCN, and tetraethyloxysilane (TEOS).
For example, the back insulating layer 164 may include a first insulating layer 164a and a second insulating layer 164b (also referred to as a first insulating film and a second insulating film, respectively). The first insulating layer 164a may prevent unwanted electrical connection between the back pad 154 and the semiconductor substrate 110. Additionally, the back pad 154 may be buried in the second insulating layer 164b such that the back surface thereof (upper surface) is exposed. The exposed back surface (upper surface) of the back pad 154 may have a back surface (upper surface) that is substantially flat with the back surface (upper surface) of the second insulating layer 164b. In some embodiments, the first insulating layer 164a and the second insulating layer 164b may be formed of the same material, but are not limited thereto and may be formed of different materials. For example, the first insulating layer 164a may include silicon nitride or silicon oxynitride, and the second insulating layer 164b may include silicon oxide.
A plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C) may be bonded to each other through direct bonding DB2 between the front insulating layer 162b and the back insulating layer 164. Additionally, the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C) may be electrically connected to each other through direct bonding DB1 between the front pad 152 and the back pad 154. Direct bonding DB1 may be intermetallic bonding, and direct bonding DB2 may be inter-dielectric bonding. A combination of direct bonding DB1 and direct bonding DB2 may be hybrid bonding.
Due to at least one of the direct bonding DB1 and direct bonding DB2, a conductive structure with a relatively low melting point, such as a bump or solder, may not be disposed between the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C). Generally, the conductive structures (e.g., bumps, solder) may be formed through a reflow process or a thermal compression bonding (TCB) process. A minimum width or pitch may be required to ensure reliability.
Due to at least one from among the direct bonding DB1 and the direct bonding DB2, since a plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C) may be bonded to each other without using the conductive structure (e.g., bump, solder), it may be advantageous to reduce the width or pitch of each of the front pad 152 and the back pad 154. Accordingly, the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C) may efficiently increase the integration of the front pad 152 and the back pad 154. The higher the integration of the front pad 152 and the back pad 154, the more the integration of electrical paths (e.g., paths through which at least one of a data signal, a control signal, a power signal, and a ground signal passes) of the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C) may also be increased. Accordingly, the semiconductor package 500 may be more advantageous for miniaturization and high performance. Alternatively, due to at least one from among the direct bonding DB1 and the direct bonding DB2, electrical reliability (e.g., impedance stability, equivalent series resistance reduction, signal integrity, power integrity, etc.) between the front pad 152 and the back pad 154 may be further improved.
Due to at least one from among the direct bonding DB1 and the direct bonding DB2, since the front surface and back surface of a plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C) may be in direct contact with each other, front and back leveling may become more important. The higher the flatness of the front and back, reliability of the front pad 152 and back pad 154 (e.g., performance to prevent poor contact or electrical shorts) and reliability of the front insulating layer 162b and back insulating layer 164 (e.g., performance to prevent pores or cracks) may be improved.
The conductive pattern 147 may be disposed between the interconnection structure 140 and the front pad 152. The support insulating layer 162a may be disposed between the device layer 120 and the front insulating layer 162b to surround the conductive pattern 147. The front surfaces (lower surfaces) and back surfaces (upper surfaces) of the conductive pattern 147 and the support insulating layer 162a may respectively be coplanar.
When the front pad 152 is coupled to the back pad 154 through an annealing process, not only the front pad 152 but also the conductive pattern 147 may be thermally expanded. Thermal expansion of the conductive pattern 147 may support thermal expansion of the back pad 154 and the front pad 152. Accordingly, the front pad 152 and the back pad 154 may be more efficiently coupled to each other. The higher the coupling efficiency between the front pad 152 and the back pad 154, the smaller the minimum volume required for the front pad 152 and the back pad 154 may be. Accordingly, the width or pitch of the front pad 152 and the back pad 154 may be further refined, and the integration of the electrical paths of the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C) may also be further increased.
The support insulating layer 162a may be implemented in the same manner as the front insulating layer 162b. For example, the support insulating layer 162a may include at least one from among SiO2, SiN, SiCN, and tetraethyloxysilane (TEOS). The conductive pattern 147 may be implemented in the same manner (e.g., same material, same process) as the interconnection layer 142 of the interconnection structure 140. Accordingly, the conductive pattern 147 may be the lowermost interconnection layer of the interconnection structure 140. For example, the conductive pattern 147 may include at least one from among copper, a copper alloy, aluminum, and an aluminum alloy.
The conductive pattern 147 may have a thickness greater than the thickness of each of the plurality of interconnection layers 142. Since the conductive pattern 147 may have a thickness greater than the thickness of each of the plurality of interconnection layers 142, and the conductive pattern 147 may have a relatively large volume. The larger the volume of the conductive pattern 147, the better the thermal expansion of the front pad 152 against the back pad 154 may be supported.
According to some embodiments of the present disclosure, the conductive pattern 147 may include a conductive material (e.g., aluminum) having a higher thermal expansion coefficient than that of the conductive material (e.g., copper) of the front pad 152. Since the thermal expansion coefficient of aluminum is higher than the thermal expansion coefficient of copper, the conductive pattern 147 containing aluminum may more efficiently support the thermal expansion of the front pad 152 containing copper.
The efficiency with which the conductive pattern 147 supports the thermal expansion bond between the front pad 152 and the back pad 154 may be a trade-off with the flatness (or importance of flatness) of the front surface of the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3, and 100C). The semiconductor package 500 and a manufacturing method thereof, according to an example embodiment, may secure one of the thermal expansion bond support efficiency and flatness (or the importance of flatness) while improving the other.
The structure of the semiconductor package according to a method of manufacturing a semiconductor package according to an example embodiment may be formed sequentially from the structure of
Referring to
Referring to
Referring to
Referring again to
Referring to
Enlarged regions A7a, A7b, A7c, and A7d of
Referring to
Referring to
Referring to
For example, the front (lower) and back (upper) surfaces of the pad pattern 147a may be polygonal or circular. Referring to
Referring to
The dummy patterns 147b and 147c may be filled in most of the spaces surrounding the pad pattern 147a to prevent the metal material from concentrating at specific points in the combination of the conductive pattern 147 and the support insulating layer 162a. Accordingly, even if the conductive pattern 147 is thicker than the interconnection layer 142, the combination of the conductive pattern 147 and the support insulating layer 162a may be stably stacked on the front (lower surface) of the device layer 120, and the occurrence of delamination of the above combination may be reliably prevented.
The dummy patterns 147b and 147c may be electrically connected to power or ground. Since the dummy patterns 147b and 147c may surround the pad pattern 147a, the pad pattern 147a may be electromagnetically shielded. Additionally, the relatively large volume of the dummy patterns 147b and 147c may improve the electrical stability of the power source or ground.
On the other hand, referring to
The upper dummy chip 200 may have a thickness T2 greater than a thickness T1a of the semiconductor chips 100A1, 100A2, and 100A3 and a thickness T1b of the semiconductor chip 100C. For example, the thickness T2 of the upper dummy chip 200 may be 200 μm or more, and the thickness T1a and the thickness T1b) may be 100 μm or less.
A lower bonding insulating layer 210 may disposed on a lower surface of the upper dummy chip 200, and a back insulating layer 174 may be disposed on an upper surface of the semiconductor chip 100C. The upper dummy chip 200 may be bonded to the back (upper surface) of the semiconductor chip 100C by directly bonding the lower bonding insulating layer 210 and the back insulating layer 174. In this manner, the upper dummy chip 200 and the semiconductor chip 100C may be bonded by inter-dielectric bonding of the lower bonding insulating layer 210 and the back insulating layer 174. At least one from among the lower bonding insulating layer 210 and the back insulating layer 174 may include a dielectric layer formed by a deposition process. In contrast, it may include a natural oxide film formed during a high temperature annealing process.
The width W2 (or area) of the upper dummy chip 200 may be equal to the width W1a (or area) of the semiconductor chips 100A1, 100A2, and 100A3, and may be the same as the width W1b (or area) of the semiconductor chip 100C, but is not limited thereto.
Referring to
An upper surface 200T of the upper dummy chip 200 is exposed by the upper surface 180T of the molding part 180. The upper surface 200T of the upper dummy chip 200, that is exposed, may have a substantially flat coplanar surface with the upper surface 180T of the molding part 180. These coplanar upper surfaces may be understood as upper surfaces obtained through a polishing process. Additionally, the side surface of the molding part 180 may have a substantially flat coplanar surface with the side surface of the base structure 300. These coplanar side surfaces may be understood as side surfaces obtained by the same cutting process.
Referring to
In this embodiment, the base structure 300 may have a width (or area) greater than the widths W1a and W1b (or area) of the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, 100A3 and 100C). The lowermost semiconductor chip (e.g., the semiconductor chip 100A1) among the plurality of semiconductor chips (e.g., the semiconductor chips 100A1, 100A2, and 100A3) may be directly bonded to the base structure 300, similar to the direct bonding DB1 and the direct bonding DB2 described above.
In detail, the front pads 152 of the semiconductor chip 100A1 adjacent to the base structure 300 may be directly bonded to the upper connection pads 354 to form direct bonding DB1. This direct bonding DB1 may bond the base structure 300 and the semiconductor chip 100A1 to each other and ensure an electrical connection. An upper bonding insulating layer 364 is formed on the upper surface of the base structure 300 employed in this embodiment, and the upper bonding insulating layer 364 may have a substantially flat upper surface that is coplanar with upper surfaces the upper connection pads 354. The upper bonding insulating layer 364 of the base structure 300 and the front insulating layer 162 of the lowermost semiconductor chip 100A1 may be directly bonded to form direct bonding DB2. In this manner, the base structure 300 and the lowermost semiconductor chip 100A1 may be hybrid bonded.
For example, the base structure 300 may be an interposer for redistribution. The base structure 300 may include a substrate body 310 and an interconnection circuit connecting the lower connection pads 352 and the upper connection pads 354 within the substrate body 310. Connection bumps 370 may be attached to the lower connection pads 352 of the base structure 300. The connection bump 370 may be, for example, a solder ball or a conductive bump. The connection bump 370 may be electrically connected to the semiconductor package 500 and a printed circuit board such as a motherboard. In some embodiments, the base structure 300 may include a semiconductor chip.
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
As set forth above, in a semiconductor package and a method of manufacturing a semiconductor package according to example embodiments, electrical characteristics and/or reliability of a front surface of a semiconductor chip may be improved.
While example embodiments have been illustrated and described above, it will be apparent to those skilled in the art that modifications and variations could be made without departing from the scope of the present disclosure.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10-2023-0144560 | Oct 2023 | KR | national |