A semiconductor structure and a method of fabricating the same
By optimizing the design and arrangement of conductive bumps and through-silicon vias, the problem of increased parasitic parameters in high-bandwidth memory chips was solved, achieving more efficient signal transmission.
Patent Information
- Application Number
- CN202211138703.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-09-19
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2042-09-19
AI Technical Summary
The increased integration density of high-bandwidth memory (HBM) chips leads to an increase in parasitic parameters in the interconnects, which affects signal transmission quality.
A small-sized first conductive structure, including conductive bumps and through-silicon vias, is used. By adjusting its material and arrangement, parasitic parameters are reduced and the signal transmission path is optimized.
It effectively reduces parasitic inductance, capacitance, and resistance, improves signal integrity and transmission efficiency, and reduces signal delay.
Smart Images

Figure CN117790444B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of semiconductor technology, and in particular to a semiconductor structure and its fabrication method. Background Technology
[0002] Typically, high-bandwidth memory (HBM) chips are stacked on the top surface of a package substrate. The HBM chip is electrically connected to the package substrate via conductive bumps. The development of 3D packaging stacking technology, driven by the demand for high bandwidth and low power consumption, has led to higher chip stacking and denser through-silicon via (TSV) interconnects. However, higher HBM integration density results in larger parasitic parameters in the interconnects. Summary of the Invention
[0003] In view of this, the present disclosure provides a semiconductor structure and a method for preparing the same.
[0004] According to a first aspect of the present disclosure, a semiconductor structure is provided, comprising:
[0005] substrate;
[0006] A chip stack is located on the substrate; the chip stack includes a plurality of chips stacked sequentially along a plane perpendicular to the substrate; wherein, each chip includes a first sub-part and a second sub-part, the first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes from the second surface of the second sub-part; the first surface and the second surface are disposed opposite to each other;
[0007] The first conductive structure includes a first conductive bump and a first through-silicon via (TSV). The first conductive bump is located between two adjacent first sub-parts of the chips. The first TSV extends through the first sub-part in a direction perpendicular to the plane of the substrate and is connected to the first conductive bump.
[0008] The first conductive bump and the first through-silicon via are made of the same material.
[0009] In some embodiments, it also includes:
[0010] A connecting bump is located between the second sub-parts of two adjacent chips.
[0011] In some embodiments, the volume of the first conductive bump is smaller than the volume of the connecting bump.
[0012] In some embodiments, the ratio of the height of the first sub-part to the height of the second sub-part in a direction perpendicular to the plane of the substrate ranges from 1.5 to 2.
[0013] In some embodiments, the first conductive bump includes a first sub-conductive bump and a second sub-conductive bump, wherein the first sub-conductive bump and the second sub-conductive bump are respectively connected to the first sub-parts of two adjacent chips.
[0014] In some embodiments, the materials of the first conductive bump and the first through-silicon via include copper.
[0015] In some embodiments, the first conductive bump includes at least one concave surface, and the concave surfaces on adjacent first conductive bumps are disposed opposite each other.
[0016] In some embodiments, the first conductive structure is arranged in a square pattern, and in each of the multiple first conductive structures arranged in a square pattern, the concave surfaces of the first conductive bumps of the two first conductive structures at the diagonal positions are arranged opposite each other.
[0017] In some embodiments, the distance from the intersection of the diagonals of each of the four square arrangements to the concave surface of each of the first conductive bumps is a first distance, and the distance from the concave surface of the first conductive bump to the center of the first conductive bump is a second distance, and the ratio of the first distance to the second distance is 5:3 to 5:2.
[0018] In some embodiments, each chip includes n first conductive structures, where n is greater than or equal to 2;
[0019] In the projection along the plane perpendicular to the substrate, the projections of corresponding first through-silicon vias in two adjacent chip layers do not overlap.
[0020] According to a second aspect of the present disclosure, a method for fabricating a semiconductor structure is provided, comprising:
[0021] Provide substrate;
[0022] A chip stack is formed on the substrate; the chip stack includes a plurality of chips stacked sequentially along a plane perpendicular to the substrate; wherein, each chip includes a first sub-part and a second sub-part, the first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes from the second surface of the second sub-part; the first surface and the second surface are disposed opposite to each other;
[0023] After each chip is formed, a first conductive structure is formed; the first conductive structure includes a first conductive bump and a first through-silicon via (TSV). The first conductive bump is located between first sub-parts of two adjacent chips, and the first TSV extends through the first sub-part along a direction perpendicular to the plane of the substrate and is connected to the first conductive bump; wherein the first conductive bump and the first TSV are made of the same material.
[0024] In some embodiments, the first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes beyond the second surface of the second sub-part, including:
[0025] Forming the initial chip;
[0026] The second surface of the initial chip is etched to remove a portion of the initial chip. The unetched portion forms a first sub-part, and the etched portion forms a second sub-part. The first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes from the second surface of the second sub-part.
[0027] In some embodiments, it also includes:
[0028] After each chip is formed, a connecting bump is formed on the second sub-part of the chip.
[0029] In some embodiments, the volume of the first conductive bump is smaller than the volume of the connecting bump.
[0030] In some embodiments, the first conductive bump includes a first sub-conductive bump and a second sub-conductive bump, wherein the first sub-conductive bump and the second sub-conductive bump are respectively connected to the first sub-parts of two adjacent chips.
[0031] In some embodiments, the materials of the first conductive bump and the first through-silicon via include copper.
[0032] In this embodiment, because the second surface of the first sub-part of the chip protrudes beyond the second surface of the second sub-part, the distance between the first sub-parts is shortened. This allows for the use of a smaller first conductive structure, thereby reducing RLC parasitic parameters and signal delay. Simultaneously, the first conductive bump and the first through-silicon via (TSV) are made of the same material. This reduces signal reflection and distortion caused by impedance mismatch when the signal enters the first conductive bump from the TSV, thus improving signal integrity. Attached Figure Description
[0033] To more clearly illustrate the technical solutions in the embodiments of this disclosure or in the conventional art, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0034] Figure 1 This is a schematic diagram of the structure of the first conductive structure provided in an embodiment of the present disclosure;
[0035] Figure 2This is a schematic diagram of the structure of the first conductive structure provided in another embodiment of the present disclosure;
[0036] Figure 3 This is a schematic diagram of the semiconductor structure provided in the embodiments of this disclosure;
[0037] Figure 4 This is a schematic diagram showing the connection between two adjacent chip layers via a first interconnect line.
[0038] Figure 5 A flowchart illustrating a method for fabricating a semiconductor structure according to embodiments of this disclosure;
[0039] Figures 6a to 8 This is a schematic diagram of the semiconductor structure provided in the embodiments of this disclosure during the fabrication process.
[0040] Explanation of reference numerals in the attached figures:
[0041] 10-Substrate;
[0042] 20 - Chip stack; 21 - Chip; 211 - First sub-section; 212 - Second sub-section; 200 - Initial chip;
[0043] 30 - First conductive structure; 31 - First conductive bump; 301 - Concave surface; 302 - Convex surface; 32 - First through-silicon via; 300 - Initial first conductive structure; 310 - Initial first conductive bump; 31a - First sub-conductive bump; 31b - Second sub-conductive bump;
[0044] 40 - Second conductive structure; 41 - Second conductive bump;
[0045] 50 - Connecting bump;
[0046] 61 - First mask layer;
[0047] 71 - First interconnect. Detailed Implementation
[0048] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the specific embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
[0049] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of this disclosure. However, it will be apparent to those skilled in the art that this disclosure may be practiced without one or more of these details. In other instances, to avoid confusion with this disclosure, certain technical features well-known in the art have not been described; that is, not all features of actual embodiments are described herein, nor are well-known functions and structures described in detail.
[0050] In the accompanying drawings, for clarity, the dimensions of layers, areas, and elements, as well as their relative dimensions, may be exaggerated. The same reference numerals denote the same elements throughout.
[0051] It should be understood that when an element or layer is referred to as "on," "adjacent to," "connected to," or "coupled to" other elements or layers, it may be directly on, adjacent to, connected to, or coupled to other elements or layers, or there may be intervening elements or layers. Conversely, when an element is referred to as "directly on," "directly adjacent to," "directly connected to," or "directly coupled to" other elements or layers, there are no intervening elements or layers. It should be understood that although the terms first, second, third, etc., may be used to describe various elements, components, areas, layers, and / or portions, these elements, components, areas, layers, and / or portions should not be limited by these terms. These terms are only used to distinguish one element, component, area, layer, or portion from another element, component, area, layer, or portion. Therefore, without departing from the teachings of this disclosure, the first element, component, area, layer, or portion discussed below may be referred to as a second element, component, area, layer, or portion. And the discussion of a second element, component, area, layer, or portion does not imply that the first element, component, area, layer, or portion necessarily exists in this disclosure.
[0052] Spatial relation terms such as “below,” “under,” “below,” “under,” “above,” “above,” etc., are used herein for convenience of description to describe the relationship between one element or feature shown in the figure and other elements or features. It should be understood that, in addition to the orientation shown in the figure, spatial relation terms are intended to also include different orientations of the device in use and operation. For example, if the device in the figure is flipped, then the element or feature described as “below,” “under,” or “below” other elements or features will be oriented “above” other elements or features. Therefore, the exemplary terms “below” and “under” can include both above and below orientations. The device may be otherwise oriented (rotated 90 degrees or otherwise) and the spatial descriptive terms used herein will be interpreted accordingly.
[0053] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. When used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprise” and / or “comprising,” when used in this specification, identify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups. When used herein, the term “and / or” includes any and all combinations of the associated listed items.
[0054] To fully understand this disclosure, detailed steps and structures will be presented in the following description to illustrate the technical solutions of this disclosure. Preferred embodiments of this disclosure are described in detail below; however, other embodiments may also be implemented in addition to these detailed descriptions.
[0055] This disclosure provides a semiconductor structure.
[0056] Figure 1 This is a schematic diagram of the first conductive structure provided in an embodiment of the present disclosure.
[0057] like Figure 1 As shown, the first conductive structure 30 includes a first conductive bump 31, the first conductive bump 31 includes at least one concave surface 301, and the concave surfaces 301 on adjacent first conductive bumps 31 are disposed opposite to each other.
[0058] In this embodiment, when a signal passes through one of the first conductive bumps, the edge field radiation effect causes parasitic RLC to be introduced into other surrounding first conductive bumps, and this RLC is inversely proportional to the distance; the farther the distance, the weaker the edge field radiation effect. Therefore, by setting the concave surfaces of adjacent first conductive bumps opposite each other, the overlap range of the edge field in space is reduced, thereby reducing the parasitic parameters caused by edge field radiation. Simultaneously, by setting the first conductive bump to include at least one concave surface, the volume of the first conductive bump is reduced, thereby reducing the parasitic capacitance of the first conductive bump itself.
[0059] In one embodiment, the first conductive structure 30 is arranged in a square pattern, and in each of the multiple first conductive structures 30 arranged in a square pattern, the concave surfaces 301 of the first conductive bumps 31 of the two first conductive structures 30 at the diagonal positions are arranged opposite each other.
[0060] In this embodiment, the concave surfaces of the first conductive bumps at the diagonal positions are arranged opposite each other. This increases the distance between the first conductive bumps, thereby reducing the edge field between the first conductive bumps and thus reducing the RLC parasitic parameters.
[0061] In some implementations, such as Figure 1 As shown, the first conductive structures 30 are arranged in a square, that is, the four conductive structures form a rectangle. In some other embodiments, the four first conductive structures may also be formed into a rhombus or trapezoidal shape.
[0062] In one embodiment, see Figure 1 The first conductive bump 31 further includes at least one convex surface 302, which is disposed adjacent to the concave surface 301. By providing the convex surface, it is easier to perform subsequent bonding of the first conductive structure, thus ensuring the bonding quality of the first conductive bump 31.
[0063] like Figure 1 As shown, in each of the multiple first conductive structures 30 arranged in a square, the first conductive bump 31 of each first conductive structure 30 includes multiple concave surfaces 301, and a convex surface 302 is provided between two adjacent concave surfaces 301. The area of the concave surface 301 is larger than the area of the convex surface 302.
[0064] The concave surface is designed to increase the distance between the two first conductive bumps, thereby reducing RLC parasitic parameters. Therefore, the area of the concave surface is set to be larger to facilitate the reduction of parasitic parameters. The convex surface is designed to facilitate bonding, since the area of the convex surface does not need to be set too large, it is only necessary to facilitate bonding.
[0065] In some embodiments, each of the first conductive bumps includes a concave surface, which is disposed opposite to the center position of each of the four square bumps. In this embodiment, each first conductive bump has only one concave surface, but this concave surface is disposed opposite to the concave surfaces of the other first conductive bumps. Therefore, parasitic parameters can be reduced to a certain extent, and process steps can be reduced, thus lowering manufacturing costs.
[0066] In other embodiments, the first conductive bump includes two concave surfaces, with the concave surfaces of two adjacent first conductive bumps arranged opposite each other. In this embodiment, two concave surfaces are provided, which can further reduce parasitic parameters. At the same time, because only two concave surfaces are provided, the area of the convex surface is relatively large, increasing the bonding area and ensuring the quality of bonding.
[0067] In other embodiments, such as Figure 1 As shown, each first conductive structure 30 has a first conductive bump 31 including a plurality of concave surfaces 301.
[0068] In one embodiment, such as Figure 1As shown, the distance from the intersection of the diagonals of each square arrangement to the concave surface 301 of each of the first conductive bumps 31 is the first distance h1, and the distance from the concave surface 301 of the first conductive bump 31 to the center of the first conductive bump 31 is the second distance h2. The ratio of the first distance h1 to the second distance h2 is 5:3 to 5:2.
[0069] If the ratio of the first distance to the second distance is set too large, it means that the concave surface of the first conductive bump is too close to the center of the first conductive bump, which will result in an excessively small area of the first conductive bump, affecting its conductivity. Conversely, if the ratio of the first distance to the second distance is set too small, it means that the concave surface of the first conductive bump is close to the intersection of the diagonals, which reduces the distance between adjacent first conductive bumps and increases parasitic parameters. Therefore, setting the ratio of the first distance to the second distance to 5:3 to 5:2 ensures the conductivity of the first conductive bump while reducing parasitic parameters.
[0070] In one embodiment, the first conductive structure 30 further includes a first through-silicon via (TSV) 32 located on the first conductive bump 31. The TSV 32 and the first conductive bump 31 ensure electrical connections between subsequent substrates and chip stacks, as well as between adjacent chips.
[0071] The conductive material inside the first through-silicon via 32 includes, but is not limited to, Cu, and the conductive material is wrapped with an insulating material, which includes, but is not limited to, SiO2.
[0072] In one embodiment, such as Figure 2 As shown, the semiconductor structure further includes: a second conductive structure 40, which is located at the intersection of the diagonals of each of the four-sided arrangement; the second conductive structure 40 includes a second conductive bump 41, which includes at least one concave surface.
[0073] A second conductive structure 40 is added in the middle of the first conductive structure 30 arranged in a square. The first conductive structure 30 is a signal conductive structure, that is, the first conductive structure 30 transmits high voltage signals, and the second conductive structure 40 is a ground conductive structure, that is, the second conductive structure 40 transmits low voltage signals. Since the signal will choose the nearest ground or power source as the return path during the transmission process, and the second conductive structure is closer to the first conductive structure, the capacity of the electromagnetic flow to the ground conductive structure, that is, the second conductive structure, increases, and the capacity flowing to the first conductive structure will be relatively reduced. This can effectively reduce the edge field effect and thus reduce the RLC parasitic parameters in the return path segment.
[0074] In one embodiment, each concave surface of the second conductive bump 41 is disposed opposite to one concave surface of the adjacent first conductive bump 31. The opposing concave surfaces of the second and first conductive bumps increase the distance between them, thereby reducing crosstalk and parasitic parameters.
[0075] In one embodiment, such as Figure 4 As shown, the first conductive structure can be used in a multi-chip stacked structure to electrically connect adjacent chips and improve the connection method to further reduce RLC parasitic parameters.
[0076] Specifically, in Figure 4 In the process, two adjacent chip layers 21 can be electrically connected through the first conductive bump 31 and the first through-silicon via 32. Then, for... Figure 4 The connection method between adjacent chips in the chip stacking structure shown will be further described.
[0077] In one embodiment, such as Figure 3 and Figure 4 As shown, the chip stack 20 includes a plurality of chips 21 stacked sequentially, and each chip 21 includes n first conductive structures 30, where n is greater than or equal to 2.
[0078] In the projection along the plane direction perpendicular to the substrate 10, the projections of the corresponding first through-silicon vias 32 in two adjacent chip layers 21 do not overlap.
[0079] In this embodiment, the projections of the first through-silicon vias (TSVs) of corresponding first conductive structures within adjacent chip layers do not overlap, indicating that the corresponding TSVs within adjacent chip layers are offset at a certain angle. This allows the same signal to rotate and rise within the structure formed by stacked chips, reducing crosstalk between different signals. Simultaneously, the optimized spatial structure enables the formation of a higher bandwidth memory.
[0080] In one embodiment, such as Figure 3 As shown, the semiconductor structure further includes a first interconnect 71, through which corresponding first conductive structures 30 in adjacent layers of the chip 21 are connected. By forming the first interconnect 71 in the chip, the connection of the first conductive structures 30 when spirally arranged is achieved, thereby ensuring normal signal transmission.
[0081] Specifically, such as Figure 3As shown, each chip layer may include multiple first conductive structures 30, namely CH0, CH1, CH2, and CH3. The corresponding CH0 within each chip layer is connected via a first interconnect 71 and rotates upwards at a certain angle. Similarly, CH1, CH2, and CH3 are also connected via first interconnect 71 and rotate upwards at a certain angle. The first interconnects connecting corresponding first conductive structures within the same layer are deflected at a certain angle, thus reducing the area of the first interconnects facing each other and consequently reducing crosstalk between the first interconnects.
[0082] In one embodiment, one end of the first interconnect is connected to the first through-silicon via, and the other end of the first interconnect is connected to the first conductive bump.
[0083] The first interconnect is a metal wire, such as Figure 3 As shown, it includes metal wires M0 to M4.
[0084] like Figure 3 As shown, one end of the first interconnect 71 is M0, which is connected to the first through-silicon via (TSV) or the first conductive bump of the first conductive structure within one of the chip layers. The other end, M4, is connected to the first conductive bump or the first TSV of the corresponding first conductive structure within an adjacent chip. That is, if one end is connected to the first TSV, the other end is connected to the first conductive bump, and vice versa. M0 and M4 are connected by M1, M2, and M3.
[0085] It needs to be explained that, Figure 3 The end face of CH0 in the lower-level chip connected by M0 and the end face of CH0 in the upper-level chip connected by M4 should be on the same horizontal plane, that is, the first interconnect line should be parallel to the plane of the chip. Specifically, see [link to relevant documentation]. Figure 4 The signal is transmitted in the direction of the arrow, from the first conductive structure in one layer of the chip to the corresponding first conductive structure in the adjacent layer of the chip, wherein the first interconnect is located at the position indicated by the arrow parallel to the plane of the chip.
[0086] like Figure 3As shown, the first conductive structures 30 (e.g., CH0 and CH1) within the semiconductor structure are arranged spirally in the stacking direction, rather than vertically. This means that the distance between the same first conductive structure 30 (e.g., CH0) in adjacent chip layers increases. If the first conductive structures 30 (e.g., CH0 and CH1) within the semiconductor structure were arranged vertically, all first conductive structures 30 within the chip would generate signal crosstalk due to edge field effects. Furthermore, because the first conductive structures 30 are vertically arranged, the same first conductive structures 30 in adjacent chips are closer together, leading to a superposition of crosstalk effects. Moreover, the longer the signal formed by the first conductive structure 30, the stronger the superposition of crosstalk effects, ultimately causing signal distortion in the top-layer chip.
[0087] In this embodiment, since the first conductive structures 30 (e.g., CH0 and CH1) are spirally arranged, the distance between the same first conductive structure 30 (e.g., CH0) in two adjacent chip layers will increase. As a result, when two different signals crosstalk within the same chip, the crosstalk effect will not be superimposed on the other chip, thereby improving the impact of crosstalk on the signal.
[0088] like Figure 3 As shown, the first conductive structure 30 can be a through-silicon via (TSV) structure. CH0 and CH1 can represent different TSVs, that is, TSVs that transmit different signals.
[0089] In some embodiments, the chip can be further improved to further reduce the RLC parasitic parameters of the semiconductor structure and improve performance.
[0090] In one embodiment, such as Figure 4 As shown, the semiconductor structure further includes:
[0091] substrate 10;
[0092] A chip stack 20 is disposed on the substrate 10 through a plurality of first conductive structures 30; the chip stack 20 includes a plurality of chips 21 stacked sequentially along a direction perpendicular to the plane of the substrate 10; wherein, the chip 21 includes a first sub-part 211 and a second sub-part 212, the first surface of the first sub-part 211 and the first surface of the second sub-part 212 are flush, and the second surface of the first sub-part 211 protrudes from the second surface of the second sub-part 212; the first surface and the second surface are disposed opposite to each other.
[0093] In one embodiment, the substrate 10 may be a printed circuit board (PCB), a redistributed substrate, or a logic chip.
[0094] like Figure 4As shown, the first conductive bump 31 of the first conductive structure 30 is located between the first sub-parts 211 of two adjacent chips 21, and the first through-silicon via 32 extends through the first sub-part 211 along a direction perpendicular to the plane of the substrate 10 and is connected to the first conductive bump 31; wherein, the first conductive bump 31 and the first through-silicon via 32 are made of the same material.
[0095] In this embodiment, because the second surface of the first sub-part of the chip protrudes beyond the second surface of the second sub-part, the distance between the first sub-parts is shortened. This allows for the use of a smaller first conductive structure, thereby reducing RLC parasitic parameters and signal delay. Simultaneously, the first conductive bump and the first through-silicon via (TSV) are made of the same material. This reduces signal reflection and distortion caused by impedance mismatch when the signal enters the first conductive bump from the TSV, thus improving signal integrity.
[0096] Specifically, in Figure 4 In the embodiment shown, the first conductive bump 31 includes a first sub-conductive bump 31a and a second sub-conductive bump 31b, and the first sub-conductive bump 31a and the second sub-conductive bump 31b are respectively connected to the first sub-parts 211 of two adjacent chips 21.
[0097] The first sub-conductive bump and the second sub-conductive bump can be conductive pads.
[0098] In one embodiment, the materials of the first conductive bump 31 and the first through-silicon via 32 include, but are not limited to, copper (Cu).
[0099] In this embodiment, small-sized copper conductive pads are used to bond different chips, replacing the traditional uBump bonding method. This can significantly reduce the RLC parasitic parameters introduced by uBump bonding and reduce signal delay.
[0100] In one embodiment, the structure of the first conductive bump in this embodiment can be referred to Figure 1 The structure of the first conductive bump in the illustrated embodiment is formed to include at least one concave surface.
[0101] In some embodiments, it is also possible to Figure 4 A second conductive structure is set in the middle. The arrangement of the second conductive structure and the first conductive structure can be referred to Figure 2 .
[0102] The connection method of the silicon vias in adjacent chips in this embodiment can also be referred to. Figure 3 The connection method shown in the embodiment.
[0103] In one embodiment, the semiconductor structure further includes connecting bumps 50 located between the second sub-parts 212 of the chip 21. The connecting bumps do not transmit signals; they primarily support the chip and also facilitate heat dissipation of the first sub-parts.
[0104] Table 1 shows a comparison of simulated data between semiconductor structures in the prior art and semiconductor structures in the embodiments of this disclosure. It should be noted that in the semiconductor structures of the prior art, the first and second surfaces of the first and second sub-parts of the chip are flush; that is, the semiconductor structures of the prior art do not have a protruding first sub-part.
[0105] Table 1
[0106]
[0107] As can be seen from the comparison of the simulation data in Table 1, at an operating frequency of 1 GHz, the parasitic inductance L, parasitic capacitance C, and parasitic resistance R of the semiconductor structure in this embodiment are reduced by 60.71%, 17.46%, and 86.30%, respectively, compared to the semiconductor structure in the prior art. This demonstrates that the semiconductor structure provided in this embodiment can reduce the parasitic parameter RLC and improve device performance.
[0108] In one embodiment, the volume of the first conductive bump 31 is smaller than the volume of the connecting bump 50. The first conductive bump serves to transmit signals, and in order to reduce parasitic parameters, the first conductive bump is made into a small-volume structure, while the connecting bump is made into a large-volume structure in order to support the chip.
[0109] In one embodiment, the ratio of the height of the first sub-part 211 to the height of the second sub-part 212 in a direction perpendicular to the plane of the substrate 10 ranges from 1.5 to 2.
[0110] If the ratio of the height of the first sub-part to the height of the second sub-part is too large, it means that the height of the second sub-part is too small, which will result in less available space on the chip surface for device structure, affecting device performance. If the ratio is too small, the height difference between the first and second sub-parts will be small. In order to ensure the small size of the first conductive bump, the distance between the chips will be small, and the size of the connecting bump will also be reduced. This may result in poor support.
[0111] In one specific embodiment, the height of the first sub-part can be 41 μm, and the height of the second sub-part can be 25 μm.
[0112] This disclosure also provides a method for fabricating a semiconductor structure; please refer to the appendix for details. Figure 5 As shown in the figure, the method includes the following steps:
[0113] Step 501: Provide a substrate;
[0114] Step 502: Form a chip stack on the substrate; the chip stack includes a plurality of chips stacked sequentially along a plane perpendicular to the substrate; wherein, each chip includes a first sub-part and a second sub-part, the first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes from the second surface of the second sub-part; the first surface and the second surface are disposed opposite to each other;
[0115] Step 503: After forming each chip, a first conductive structure is formed; the first conductive structure includes a first conductive bump and a first through-silicon via (TSV), the first conductive bump is located between the first sub-parts of two adjacent chips, the first TSV extends through the first sub-part along a direction perpendicular to the plane of the substrate and is connected to the first conductive bump; wherein the first conductive bump and the first TSV are made of the same material.
[0116] The method for preparing the semiconductor structure provided in this disclosure will be further described in detail below with reference to specific embodiments.
[0117] Figures 6a to 8 This is a schematic diagram of the semiconductor structure provided in the embodiments of this disclosure during the fabrication process.
[0118] See first Figures 6a to 6d The preparation process of the first conductive structure is explained in detail.
[0119] See Figures 6a to 6c The formation of the first conductive structure 30 includes:
[0120] An initial first conductive structure 300 is formed, the initial first conductive structure 300 includes an initial first conductive bump 310, the initial first conductive bump 310 being circular in shape;
[0121] At least one first mask layer 61 is formed on each of the initial first conductive bumps 310, and the first mask layer 61 covers a portion of the periphery of the initial first conductive bump 310;
[0122] The portion of the initial first conductive bump 310 covered by the first mask layer 61 is etched away to form the first conductive structure 30.
[0123] In one embodiment, the first mask layer 61 is circular in shape, such that the first conductive bump formed after the initial first conductive bump 310 is partially removed includes at least one concave surface.
[0124] It is understandable that the first mask layer may also be other arc-shaped structures.
[0125] In one embodiment, the first conductive structure 30 is arranged in a square pattern, and in each of the multiple first conductive structures 30 arranged in a square pattern, the concave surfaces 301 of the first conductive bumps 31 of the two first conductive structures 30 at the diagonal positions are arranged opposite each other.
[0126] In this embodiment, the concave surfaces of the first conductive bumps at the diagonal positions are arranged opposite each other. This increases the distance between the first conductive bumps, thereby reducing the edge field between the first conductive bumps and thus reducing the RLC parasitic parameters.
[0127] In some implementations, such as Figure 6c As shown, the first conductive structures 30 are arranged in a square, that is, the four conductive structures form a rectangle. In some other embodiments, the four first conductive structures may also be formed into a rhombus or trapezoidal shape.
[0128] In one embodiment, see Figure 6c The first conductive bump 31 further includes at least one convex surface 302, which is disposed adjacent to the concave surface 301. By providing the convex surface, it is easier to perform subsequent bonding of the first conductive structure, thus ensuring the bonding quality of the first conductive bump 31.
[0129] like Figure 6c As shown, in each of the multiple first conductive structures 30 arranged in a square, the first conductive bump 31 of each first conductive structure 30 includes multiple concave surfaces 301, and a convex surface 302 is provided between two adjacent concave surfaces 301. The area of the concave surface 301 is larger than the area of the convex surface 302.
[0130] The concave surface is designed to increase the distance between the two first conductive bumps, thereby reducing RLC parasitic parameters. Therefore, the area of the concave surface is set to be larger to facilitate the reduction of parasitic parameters. The convex surface is designed to facilitate bonding, since the area of the convex surface does not need to be set too large, it is only necessary to facilitate bonding.
[0131] In some embodiments, each of the first conductive bumps includes a concave surface, which is disposed opposite to the center position of each of the four square bumps. In this embodiment, each first conductive bump has only one concave surface, but this concave surface is disposed opposite to the concave surfaces of the other first conductive bumps. Therefore, parasitic parameters can be reduced to a certain extent, and process steps can be reduced, thus lowering manufacturing costs.
[0132] In other embodiments, the first conductive bump includes two concave surfaces, with the concave surfaces of two adjacent first conductive bumps arranged opposite each other. In this embodiment, two concave surfaces are provided, which can further reduce parasitic parameters. At the same time, because only two concave surfaces are provided, the area of the convex surface is relatively large, increasing the bonding area and ensuring the quality of bonding.
[0133] In other embodiments, such as Figure 6c As shown, each first conductive structure 30 has a first conductive bump 31 including a plurality of concave surfaces 301.
[0134] In one embodiment, such as Figure 6c As shown, the distance from the intersection of the diagonals of each square arrangement to the concave surface 301 of each of the first conductive bumps 31 is the first distance h1, and the distance from the concave surface 301 of the first conductive bump 31 to the center of the first conductive bump 31 is the second distance h2. The ratio of the first distance h1 to the second distance h2 is 5:3 to 5:2.
[0135] If the ratio of the first distance to the second distance is set too large, it means that the concave surface of the first conductive bump is too close to the center of the first conductive bump, which will result in an excessively small area of the first conductive bump, affecting its conductivity. Conversely, if the ratio of the first distance to the second distance is set too small, it means that the concave surface of the first conductive bump is close to the intersection of the diagonals, which reduces the distance between adjacent first conductive bumps and increases parasitic parameters. Therefore, setting the ratio of the first distance to the second distance to 5:3 to 5:2 ensures the conductivity of the first conductive bump while reducing parasitic parameters.
[0136] In one embodiment, the first conductive structure 30 further includes a first through-silicon via (TSV) 32 located on the first conductive bump 31. The TSV 32 and the first conductive bump 31 ensure electrical connections between subsequent substrates and chip stacks, as well as between adjacent chips.
[0137] The conductive material inside the first through-silicon via 32 includes, but is not limited to, Cu, and the conductive material is wrapped with an insulating material, which includes, but is not limited to, SiO2.
[0138] Next, see Figure 6d The method further includes forming a second conductive structure 40 at the intersection of each diagonal line arranged in a square, the second conductive structure 40 including a second conductive bump 41, the second conductive bump 41 including at least one concave surface.
[0139] A second conductive structure 40 is added in the middle of the first conductive structure 30 arranged in a square. The first conductive structure 30 is a signal conductive structure, that is, the first conductive structure 30 transmits high voltage signals, and the second conductive structure 40 is a ground conductive structure, that is, the second conductive structure 40 transmits low voltage signals. Since the signal will choose the nearest ground or power source as the return path during the transmission process, and the second conductive structure is closer to the first conductive structure, the capacity of the electromagnetic flow to the ground conductive structure, that is, the second conductive structure, increases, and the capacity flowing to the first conductive structure will be relatively reduced. This can effectively reduce the edge field effect and thus reduce the RLC parasitic parameters in the return path segment.
[0140] In one embodiment, each concave surface of the second conductive bump 41 is disposed opposite to one concave surface of the adjacent first conductive bump 31. The opposing concave surfaces of the second and first conductive bumps increase the distance between them, thereby reducing crosstalk.
[0141] Next, see Figures 7a to 7d The first conductive structure prepared by the above method can be applied to the chip stack, which can further improve the chip, thereby further reducing the RLC parasitic parameters of the semiconductor structure and improving performance.
[0142] See first Figure 7a Step 501 is executed, providing substrate 10.
[0143] In one embodiment, the substrate 10 may be a printed circuit board (PCB), a redistributed substrate, or a logic chip.
[0144] Next, see Figures 7b to 7d Steps 502 and 503 are executed to form a chip stack 20 on the substrate 10. The chip stack 20 includes a plurality of chips 21 stacked sequentially along a direction perpendicular to the plane of the substrate 10. Each chip 21 includes a first sub-part 211 and a second sub-part 212. The first surface of the first sub-part 211 and the first surface of the second sub-part 212 are flush, and the second surface of the first sub-part 211 protrudes from the second surface of the second sub-part 212. The first surface and the second surface are disposed opposite to each other.
[0145] After each chip 21 is formed, a first conductive structure 30 is formed; the first conductive structure 30 includes a first conductive bump 31 and a first through-silicon via 32. The first conductive bump 31 is located between the first sub-parts 211 of two adjacent chips 21. The first through-silicon via 32 extends through the first sub-parts 211 in a direction perpendicular to the plane of the substrate 10 and is connected to the first conductive bump 31; wherein the first conductive bump 31 and the first through-silicon via 32 are made of the same material.
[0146] In this embodiment, because the second surface of the first sub-part of the chip protrudes beyond the second surface of the second sub-part, the distance between the first sub-parts is shortened. This allows for the use of a smaller first conductive structure, thereby reducing RLC parasitic parameters and signal delay. Simultaneously, the first conductive bump and the first through-silicon via (TSV) are made of the same material. This reduces signal reflection and distortion caused by impedance mismatch when the signal enters the first conductive bump from the TSV, thus improving signal integrity.
[0147] Specifically, in Figure 7d In the illustrated embodiment, the first conductive bump 31 includes a first sub-conductive bump 31a and a second sub-conductive bump 31b, which are respectively connected to the first sub-parts 211 of two adjacent chips 21. The first sub-conductive bump 31a and the second sub-conductive bump 31b may also include at least one concave surface to further reduce signal crosstalk between them.
[0148] The first sub-conductive bump and the second sub-conductive bump can be conductive pads.
[0149] In one embodiment, the materials of the first conductive bump 31 and the first through-silicon via 32 include, but are not limited to, copper (Cu).
[0150] In this embodiment, small-sized copper conductive pads are used to bond different chips, replacing the traditional uBump bonding method. This can significantly reduce the RLC parasitic parameters introduced by uBump bonding and reduce signal delay.
[0151] In one embodiment, see Figure 7b and Figure 7c The first surface of the first sub-part 211 and the first surface of the second sub-part 212 are flush, and the second surface of the first sub-part 211 protrudes from the second surface of the second sub-part 212, including:
[0152] An initial chip 200 is formed; the second surface (e.g., the back side) of the initial chip 200 is etched to remove a portion of the initial chip, wherein the unetched portion is formed as a first sub-part 211 and the etched portion is formed as a second sub-part 212, the first surface of the first sub-part 211 and the first surface of the second sub-part 212 are flush, and the second surface of the first sub-part 211 protrudes from the second surface of the second sub-part 212.
[0153] In one embodiment, the method further includes forming a connecting bump 50 on a second sub-part 212 of each chip 21 after forming each chip 21. The connecting bump does not function to transmit signals, but mainly serves to support the chip and also facilitates heat dissipation of the first sub-part.
[0154] In one embodiment, the volume of the first conductive bump 31 is smaller than the volume of the connecting bump 50. The first conductive bump serves to transmit signals, and in order to reduce parasitic parameters, the first conductive bump is made into a small-volume structure, while the connecting bump is made into a large-volume structure in order to support the chip.
[0155] In one embodiment, the ratio of the height of the first sub-part 211 to the height of the second sub-part 212 in a direction perpendicular to the plane of the substrate 10 ranges from 1.5 to 2.
[0156] If the ratio of the height of the first sub-part to the height of the second sub-part is too large, it means that the height of the second sub-part is too small, which will result in less available space on the chip surface for device structure, affecting device performance. If the ratio is too small, the height difference between the first and second sub-parts will be small. In order to ensure the small size of the first conductive bump, the distance between the chips will be small, and the size of the connecting bump will also be reduced. This may result in poor support.
[0157] In one specific embodiment, the height of the first sub-part can be 41 μm, and the height of the second sub-part can be 25 μm.
[0158] In some embodiments, the connection method of the chips in the chip stack prepared by the above method can be improved to further reduce RLC parasitic parameters. Specifically, see [link to relevant documentation]. Figure 8 .
[0159] See Figure 8 Each chip 21 of the chip stack includes n first conductive structures 30, where n is greater than or equal to 2; in the projection along the plane direction perpendicular to the substrate 10, the projections of corresponding first through-silicon vias 32 in two adjacent chip layers 21 do not overlap.
[0160] See also Figure 8The method further includes: after forming each chip 21, forming a first interconnect line 71, and connecting corresponding first conductive structures 30 in two adjacent chip 21 layers through the first interconnect line 71.
[0161] Specifically, such as Figure 8 As shown, each chip layer may include multiple first conductive structures 30, namely CH0, CH1, CH2, and CH3. The corresponding CH0 within each chip layer is connected via a first interconnect 71 and rotates upwards at a certain angle. Similarly, CH1, CH2, and CH3 are also connected via first interconnect 71 and rotate upwards at a certain angle. The first interconnects connecting corresponding first conductive structures within the same layer are deflected at a certain angle, thus reducing the area of the first interconnects facing each other and consequently reducing crosstalk between the first interconnects.
[0162] In one embodiment, one end of the first interconnect is connected to the first through-silicon via, and the other end of the first interconnect is connected to the first conductive bump.
[0163] The first interconnect is a metal wire, such as Figure 8 As shown, it includes metal wires M0 to M4.
[0164] like Figure 8 As shown, one end of the first interconnect 71 is M0, which is connected to the first through-silicon via (TSV) or the first conductive bump of the first conductive structure within one of the chip layers. The other end, M4, is connected to the first conductive bump or the first TSV of the corresponding first conductive structure within an adjacent chip. That is, if one end is connected to the first TSV, the other end is connected to the first conductive bump, and vice versa. M0 and M4 are connected by M1, M2, and M3.
[0165] It needs to be explained that, Figure 8 The end face of CH0 in the lower-level chip connected by M0 and the end face of CH0 in the upper-level chip connected by M4 should be on the same horizontal plane, that is, the first interconnect line should be parallel to the plane of the chip. Specifically, see [link to relevant documentation]. Figure 7d The signal is transmitted in the direction of the arrow, from the first conductive structure in one layer of the chip to the corresponding first conductive structure in the adjacent layer of the chip. The first interconnect is located at the position indicated by the arrow, which is parallel to the plane of the chip.
[0166] The above description is merely a preferred embodiment of this disclosure and is not intended to limit the scope of protection of this disclosure. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.
Claims
1. A semiconductor structure, characterized in that, include: substrate; A chip stack is located on the substrate; the chip stack includes a plurality of chips stacked sequentially along a plane perpendicular to the substrate; wherein, each chip includes a first sub-part and a second sub-part, the first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes from the second surface of the second sub-part; the first surface and the second surface are disposed opposite to each other; The first conductive structure includes a first conductive bump and a first through-silicon via (TSV). The first conductive bump is located between two adjacent first sub-parts of the chips. The first TSV extends through the first sub-part in a direction perpendicular to the plane of the substrate and is connected to the first conductive bump. A connecting bump is located between the second sub-parts of two adjacent chips; The first conductive bump and the first through-silicon via are made of the same material.
2. The semiconductor structure according to claim 1, characterized in that, The volume of the first conductive bump is smaller than the volume of the connecting bump.
3. The semiconductor structure according to claim 1, characterized in that, Along a direction perpendicular to the plane of the substrate, the ratio of the height of the first sub-part to the height of the second sub-part ranges from 1.5 to 2.
4. The semiconductor structure according to claim 1, characterized in that, The first conductive bump includes a first sub-conductive bump and a second sub-conductive bump, and the first sub-conductive bump and the second sub-conductive bump are respectively connected to the first sub-parts of two adjacent chips.
5. The semiconductor structure according to claim 1, characterized in that, The materials of the first conductive bump and the first through-silicon via include copper.
6. The semiconductor structure according to claim 1, characterized in that, The first conductive bump includes at least one concave surface, and the concave surfaces on adjacent first conductive bumps are disposed opposite each other.
7. The semiconductor structure according to claim 6, characterized in that, The first conductive structure is arranged in a square pattern. In each square arrangement of multiple first conductive structures, the concave surfaces of the first conductive bumps of the two first conductive structures at the diagonal positions are arranged opposite each other.
8. The semiconductor structure according to claim 7, characterized in that, The distance from the intersection of the diagonals of each of the four square arrangements to the concave surface of each of the first conductive bumps is the first distance, and the distance from the concave surface of the first conductive bump to the center of the first conductive bump is the second distance. The ratio of the first distance to the second distance is 5:3 to 5:
2.
9. The semiconductor structure according to claim 7, characterized in that, Each of the chips includes n first conductive structures, where n is greater than or equal to 2; In the projection along the plane perpendicular to the substrate, the projections of corresponding first through-silicon vias in two adjacent chip layers do not overlap.
10. A method for fabricating a semiconductor structure, characterized in that, include: Provide substrate; A chip stack is formed on the substrate; the chip stack includes a plurality of chips stacked sequentially along a plane perpendicular to the substrate; wherein, each chip includes a first sub-part and a second sub-part, the first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes from the second surface of the second sub-part; the first surface and the second surface are disposed opposite to each other; After each chip is formed, a first conductive structure is formed; the first conductive structure includes a first conductive bump and a first through-silicon via (TSV). The first conductive bump is located between the first sub-parts of two adjacent chips, and the first TSV extends through the first sub-part along a direction perpendicular to the plane of the substrate and is connected to the first conductive bump; wherein the first conductive bump and the first TSV are made of the same material. In this process, after each chip is formed, a connecting bump is formed on the second sub-part of the chip.
11. The method according to claim 10, characterized in that, The first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes from the second surface of the second sub-part, including: Forming the initial chip; The second surface of the initial chip is etched to remove a portion of the initial chip. The unetched portion forms a first sub-part, and the etched portion forms a second sub-part. The first surface of the first sub-part and the first surface of the second sub-part are flush, and the second surface of the first sub-part protrudes from the second surface of the second sub-part.
12. The method according to claim 10, characterized in that, The volume of the first conductive bump is smaller than the volume of the connecting bump.
13. The method according to claim 11, characterized in that, The first conductive bump includes a first sub-conductive bump and a second sub-conductive bump, and the first sub-conductive bump and the second sub-conductive bump are respectively connected to the first sub-parts of two adjacent chips.
14. The method according to claim 11, characterized in that, The materials of the first conductive bump and the first through-silicon via include copper.
Citation Information
Patent Citations
Semiconductor package
CN107507825A
Semiconductor package and method of manufacturing the semiconductor package
US20220181299A1
