Semiconductor products with interlocked metal-to-metal bonding and methods for manufacturing them

Abstract

Semiconductor products having interlocked metal-to-metal bonding and methods thereof. The present invention provides a structure and method for implementing metal-to-metal bonding in electronic devices. For example, and without any limitation, various aspects of the present invention provide a semiconductor device utilizing an interlocking structure configured to enhance metal-to-metal bonding and a method thereof.

Description

[0001] This application is a divisional application of application number 201710012110.3 filed on January 6, 2017, entitled "Semiconductor product having interlocked metal-to-metal bonding and method of manufacturing thereof". Technical Field

[0002] This invention relates to a semiconductor product having interlocked metal-to-metal bonding and a method for manufacturing the same. Background Technology

[0003] Existing methods for forming electrical connections, such as those used in integrated circuits, all have drawbacks. For example, while soldering is common, solder has a relatively low melting point, which imposes temperature limitations on subsequent processing steps and the final product. Furthermore, solder atoms tend to migrate along copper contacts, altering their electrical and mechanical properties as the soldered contacts age. Direct metal-to-metal bonding (e.g., copper-to-copper (Cu-Cu) bonding) does not require solder; however, it has been proven that the high temperature, high pressure, and long dwell time complicate the assembly process, increase assembly costs, and add latency, making it impossible to mass-produce these bonding connections using cost-effective processes. Summary of the Invention

[0004] Various aspects of the present invention provide a semiconductor device for implementing metal-to-metal bonding in an electronic device and a method for manufacturing the same. For example, and without any limitation, various aspects of the present invention provide a semiconductor device employing an interlocking structure configured to enhance metal-to-metal bonding and a method for manufacturing the same.

[0005] A semiconductor device includes: a first substrate including a first metal contact structure containing metal, the first metal contact structure including a convex end; a second substrate including a second metal contact structure containing the metal, the second metal contact structure including a concave end; and a metal-to-metal junction between the convex end and the concave end at a junction region, the junction region being asymmetrical near the longitudinal axis of the concave end. Further, the semiconductor device includes a gap between the convex end and the concave end at a peripheral edge of at least one of the convex end and the concave end. Further, the junction region is offset toward a first side of the concave end. Further, the junction region is offset toward the first side of the concave end by at least 5% of the width of the first metal contact structure. Further, the junction region on the first side of the concave end is larger than on a second side of the concave end. Further, the junction region on the first side of the concave end is at least 5% larger than on the second side of the concave end. Further, at least one of the first metal contact structure and the second metal contact structure includes a copper pillar. Further, the first metal contact structure includes a first copper; the second metal contact structure includes a second copper; and the metal-to-metal bonding is a direct copper-to-copper bonding. Further, the first metal contact structure includes a first copper; the second metal contact structure includes a second copper; and the metal-to-metal bonding includes at least one intermediate metal between the first copper and the second copper. Further, at least one of the first substrate and the second substrate includes a semiconductor die.

[0006] A semiconductor device includes: a first substrate including a first metal contact structure containing metal, the first metal contact structure including a convex end; a second substrate including a second metal contact structure containing the metal, the second metal contact structure including a concave end; a metal-to-metal bond between the convex end and the concave end; and a gap between the convex end and the concave end at a peripheral edge of at least one of the convex end and the concave end. Further, the gap at a first portion of the peripheral edge is larger than at a second portion of the peripheral edge. Further, the first of the first metal contact structure and the second metal contact structure protrudes from the second of the first metal contact structure and the second metal contact structure. Further, the gap extends entirely near the peripheral edge. Further, the gap is the maximum gap between the convex end and the concave end.

[0007] A method of manufacturing a semiconductor device, the method comprising: providing a first substrate including a first metal contact structure containing metal, the first metal contact structure including a convex end; providing a second substrate including a second metal contact structure containing the metal, the second metal contact structure including a concave end; and pressing the convex end and the concave end together to form a metal-to-metal bond at a bonding region, the bonding region being asymmetrical near the longitudinal axis of the concave end. Further, the bonding region is offset toward a first side of the concave end. Further, the bonding region on the first side of the concave end is larger than on a second side of the concave end. Further, the first metal contact structure includes a first copper; the second metal contact structure includes a second copper; and the metal-to-metal bond is a direct copper-to-copper bond. Further, the first metal contact structure includes a first copper; the second metal contact structure includes a second copper; and the metal-to-metal bond includes at least one intermediate metal between the first copper and the second copper. Attached Figure Description

[0008] Figure 1 The illustrations show exemplary metal-to-metal bonding and bonding methods according to various aspects of the present invention.

[0009] Figure 2 The illustration shows an example of interlocking metal-to-metal bonding and bonding method according to various aspects of the present invention.

[0010] Figure 3 The diagram shown is a perspective view of an exemplary interlocking structure and engagement method according to various viewpoints of the present invention.

[0011] Figure 4 The diagram shown is a cross-sectional view of an exemplary interlocking structure and engagement method according to various viewpoints of the present invention.

[0012] Figure 5 The diagram shown is a cross-sectional view of an exemplary interlocking structure and engagement method according to various viewpoints of the present invention.

[0013] Figure 6 The diagram shown is a perspective view of an exemplary interlocking structure and engagement method according to various viewpoints of the present invention.

[0014] Figure 7 The diagram shown is a perspective view of an exemplary interlocking structure and engagement method according to various viewpoints of the present invention.

[0015] Figures 8A to 8E The diagram shown is a perspective view of each stage of a method for forming an interconnection structure according to various viewpoints of the present invention.

[0016] Figures 9A to 9CThe diagram shown is a perspective view of each stage of a method for forming an interconnection structure according to various viewpoints of the present invention. Detailed Implementation

[0017] The following discussion presents various aspects of the invention by providing examples of the invention. These examples are non-limiting, and therefore the scope of the various aspects of the invention should not be limited to any particular features of the examples provided herein. In the following discussion, phrases such as "for example" and "exemplary" are non-limiting and are generally synonymous with "for example, but without limitation."

[0018] As used in this article, "and / or" means any one or more items in a list connected by "and / or". For example, "x and / or y" means any element in the three-element set {(x), (y), (x,y)}. In other words, "x and / or y" means "one or both of x and y". In another example, "x, y, and / or z" means any element in the seven-element set {(x), (y), (z), (x,y), (x,z), (y,z), (x,y,z)}. In other words, "x, y, and / or z" means "one or more of x, y, and z".

[0019] The terminology used herein is for illustrative purposes only and is not intended to limit the scope of the invention. As used herein, unless otherwise clearly stated, the singular form is intended to include the plural form. It should be further understood that when the terms "comprising," "including," "having," and similar terms are used in this specification, they indicate the presence of the stated features, numbers, steps, operations, components, and / or devices, but do not preclude the presence of one or more other features, numbers, steps, operations, components, devices, and / or groups thereof, or even the inclusion of one or more other features, numbers, steps, operations, components, devices, and / or groups thereof.

[0020] It should be understood that although terms such as "first" and "second" are used herein to describe various components, these components should not be limited to these terms. These terms are only used to distinguish one component from another. Therefore, for example, the first component, first device, or first segment discussed below may also be referred to as the second component, second device, or second segment without departing from the teachings of this invention. Similarly, the terms "above," "below," "side," and similar terms may be used herein to distinguish one component from another in a relative manner. However, it should be understood that devices can be aligned in different ways; for example, a semiconductor device may be flipped to the side so that its "top" surface faces horizontal and its "side" surface faces vertical without departing from the teachings of this invention. In addition, this document uses "on top of" to indicate both "on top of" and "directly on top of" (for example, without any intermediate layer).

[0021] In the diagrams, for clarity, various dimensions (e.g., layer thickness, width, etc.) may be magnified. Furthermore, throughout all discussions of the various paradigms, the same component symbols are used to represent the same components.

[0022] While the discussion herein generally provides examples in the context of copper-to-copper bonding, it should be understood that the scope of the invention is not limited thereto. For example, the aspects of the invention are equally applicable to other metal-to-metal bonding (Au-to-Au gold bonding, Ag-to-Ag bonding, etc.). Additionally, various alloys (e.g., copper alloys, silver alloys, gold alloys, etc.) can also be used. It should be noted that the metal may also contain varying degrees of impurities. For example, a copper pillar can be formed essentially of 100% copper; however, it may also contain a specific percentage of impurities.

[0023] In one exemplary embodiment provided herein, a method of manufacturing a semiconductor device includes: providing a first substrate including a first copper contact structure having a dome-shaped end; providing a second substrate including a second copper contact structure having a dish-shaped end having a dish depth; mating the dome-shaped end and the dish-shaped end such that the dome-shaped end contacts the dish-shaped end at a plurality of points on the circumference of the dome-shaped end, and only at a distance between 5% and 95% of the dish depth extending into the dish-shaped cavity; and pressing the mated dome-shaped end and the dish-shaped end together to form a copper-to-copper bond. An electronic device manufactured according to this exemplary method is also provided herein.

[0024] In one exemplary embodiment provided herein, a method of manufacturing a semiconductor device includes: providing a first substrate including a first metal contact structure made of a metal, the first metal contact structure including a convex end, the convex end including a first convex portion having a first convex radius of curvature and a second convex portion having a second convex radius of curvature; providing a second substrate including a second metal contact structure made of the same metal, the second metal contact structure including a concave end, the concave end including a first concave portion having a first concave radius of curvature, wherein the first concave radius of curvature is smaller than the first convex radius of curvature and the first concave radius of curvature is larger than the second convex radius of curvature; mating the convex end and the concave end such that the second convex portion contacts the concave end at a plurality of points on the perimeter of the concave end and the first convex portion does not contact the concave end; and pressing the mated convex end and the concave end together to form a metal-to-metal bond made of the same metal. This article also provides an electronic device produced according to this exemplary method.

[0025] In one exemplary embodiment provided herein, a method of manufacturing a semiconductor device includes: providing a first substrate including a first metal contact structure made of a metal, the first metal contact structure including a convex end; providing a second substrate including a second metal contact structure made of the same metal, the second metal contact structure including a concave end, the concave end including a center point, an edge, and a center-to-edge distance along a surface of the concave end between the center point and the edge; mating the convex end and the concave end such that the convex end contacts the concave end at a plurality of points on the circumference of the convex end and only at distances from the center point greater than 10% and 90% of the center-to-edge distance; and pressing the mated convex end and the concave end together to form a metal-to-metal bond made of the same metal. An electronic device manufactured according to this exemplary method is also provided herein.

[0026] Figure 1This illustrates an exemplary metal-to-metal bonding process. As shown at reference numeral 100, two substrates 110 and 120 have copper contact (or interconnect) structures 112 and 122 to be bonded; these copper contacts may also be referred to herein as contacts. Substrates 110 and 120 may include a wide variety of features. For example, one or both of these substrates 110 and 120 may include a semiconductor die (e.g., in wafer form, in planar form, in a uniquely diced form, etc.). Additionally, one or both of these substrates 110 and 120 may include an interposer (e.g., formed on a carrier substrate (e.g., glass carrier, metal carrier, silicon carrier, etc.) which will be removed after the interposer is formed). Furthermore, one or both of these substrates 110 and 120 may include a single-layer substrate (e.g., a package substrate, motherboard, etc.). Accordingly, the scope of this invention should not be limited to the features of any particular type of substrate.

[0027] The copper contacts 112 and 122 shown in the figure are for illustrative purposes; however, they may, for example, include features of various types of interconnect structures. For example, one or both of these contacts 112 and 122 may include a metal pillar or rod (for example, having generally vertical side surfaces). Additionally, for example, one or both of these contacts 112 and 122 may include a metal pad or circuit. Furthermore, for example, one or both of these contacts 112 and 122 may also include a longitudinal length (or height) greater than the lateral width; and / or one or both of these contacts 112 and 122 may include a lateral width greater than the longitudinal length (or height). It should be noted that in various exemplary embodiments, the first contact 112 and the second contact 122 may be formed of the same metal (for example, copper, etc.); however, this is not necessarily the case. In another exemplary implementation, an intermediate metal may be formed between the first contact 112 and the second contact 122.

[0028] While many examples shown in this document depict these contacts 112 and 122 in cylindrical shapes, various other shapes can also be used. For example, one or both of these contacts 112 and 122 may have a rectangular cross-section (e.g., a cross-section cut perpendicular to a vertical axis, a cross-section cut perpendicular to a horizontal axis, etc.). Alternatively, one or both of these contacts 112 and 122 may have a generally polygonal (e.g., N sides, where N is an integer) cross-section (e.g., a cross-section cut perpendicular to a vertical axis, a cross-section cut perpendicular to a horizontal axis, etc.). Furthermore, one or both of these contacts 112 and 122 may have a circular or elliptical cross-section (e.g., a cross-section cut perpendicular to a vertical axis, a cross-section cut perpendicular to a horizontal axis, etc.).

[0029] Furthermore, while many examples shown herein demonstrate that the longitudinal length (in the vertical direction, for example, in the figures) of the first contact 112 (e.g., a column or rod, a long column or rod, etc.) is longer than that of the second contact 122 (e.g., a pad or base, a short column or rod, etc.), it should be understood that these relative lengths are illustrative rather than limiting. For example, the first contact 112 and the second contact 122 may have equal lengths, or the second contact 122 may be longer than the first contact 112, etc. Additionally, while the first contact 112 and the second contact 122 generally shown herein may have the same or similar axial width (e.g., the horizontal dimension, in the figures), it should be understood that these relative widths are illustrative rather than limiting. For example, the first contact 112 may be narrower than the second contact 122, or the second contact 122 may be narrower than the first contact 112, etc. Additionally, for example, the width of the first contact 112 (for example, the maximum width at its widest part) may be smaller than the width of the second contact 122 (for example, the maximum width at its widest part), and the width of the second contact 122 (for example, the maximum width at its widest part) may be smaller than the width of the first contact 112 (for example, the maximum width at its widest part), etc.

[0030] Although only one pair of mating interconnect structures is shown in the figure, it should be understood that many identical structures can be formed on these substrates. For example, contacts 112 and 122 shown in the figure could be one of dozens or hundreds of identical or similar structures used to electrically and / or mechanically couple substrates 110 and 120. For example, these contacts can be relatively closely spaced (e.g., a spacing of 50 micrometers or less, 30 micrometers or less, etc.).

[0031] In various embodiments, the first contact 112 and / or the second contact 122 can be formed in any variety of ways. For example, a metal pad or line can be formed on a silicon substrate and exposed via a dielectric layer (or passivation layer). It should be noted that this pad or line can be part of a multilayer signal distribution structure. One or more seed layers and / or under bump metal (UBM) layers can be deposited (e.g., sputtering, electroplating, etc.) on the pad or line, and, for example, the contacts(s) can be formed on the seed layers(s) and / or UBM layers via openings in a dielectric stencil, which can then be removed. In various embodiments, a solder cap can be formed on the contact; however, according to various aspects of the invention, the formed contact or its tip can remain bare (e.g., without flux).

[0032] Therefore, the scope of the present invention should not be limited to the specific characteristics of these interconnection points (for example, the first contact 112 and / or the second contact 122, etc.) or to any particular manner of manufacturing these contacts.

[0033] At symbol 150, substrates 110 and 120 are bonded (or mated) together so that copper contact 112 contacts copper contact 122. To form a Cu-Cu bond at the interface between these copper contacts 112 and 122, high pressure and high heat are typically applied for a long residence time. If the two mating surfaces of these copper contacts 112 and 122 were clean (e.g., completely clean), random atomic movement would eventually cause atoms to traverse the interface between the two surfaces, thus contaminating the interface, and potentially causing it to disappear completely over time. However, because atomic movement is typically random (e.g., mostly random or completely random), this is a slow process. For example, the typical pressure can be greater than 200 megapascals (200 MPa); the typical temperature can be greater than 300°C; and the typical residence time can range from one hour to several hours.

[0034] Now for reference Figure 2 This figure illustrates an exemplary interlocking metal-to-metal joint and jointing method according to various aspects of the present invention. Figure 2 The example structures and methods, or any part thereof, can be compared with other similar structures or methods discussed herein (for example, for...). Figure 1 as well as Figures 3 to 9C They share any or all of the common characteristics.

[0035] As indicated by symbol 200, the two substrates 210 and 220 have copper contacts 212 and 222 to be bonded. As discussed herein (for example, for...) Figure 1 These substrates 210 and 220 (etc.) may include any variety of features and / or may be formed in any variety of ways. As also discussed herein (for example, for...) Figure 1 These copper contacts 212 and 222 may include any variety of features and / or may be formed in any variety of ways.

[0036] As discussed in detail herein, the first copper contact 212 includes a convex (e.g., dome-shaped) end, and the second copper contact 222 includes a concave (e.g., dish-shaped) end.

[0037] For example, the first copper contact 212 can be formed using a metal electroplating process. The shape of the end of the first copper contact 212 (for example, the end that contacts the corresponding end of the second copper contact 222) can be formed to have a convex shape. For example, this shape can be achieved by adjusting various electroplating process parameters. For example, in one embodiment, a convex end can be achieved by increasing the concentration of the leveler used in the electroplating process. For example, doubling the concentration of the leveler used to obtain a generally flat end can achieve a convex end (or dome) height greater than 10% or 15% of the overall height of the contact.

[0038] For example, the second copper contact 222 can be formed using a metal electroplating process. The shape of the end of the second copper contact 222 (for example, the end that contacts the corresponding end of the first copper contact 212) can be formed to have a concave shape. For example, this shape can be achieved by adjusting the electroplating process parameters. For example, in one embodiment, a concave end can be achieved by reducing the concentration of the leveling agent used in the electroplating process. For example, by halving the concentration of the leveling agent used to obtain a generally flat end, a concave end (or dish-like shape) depth greater than 10% or 15% of the overall height of the contact can be achieved.

[0039] In one exemplary embodiment, the degree of convexity and / or concavity at each point on the ends of these contacts can be adjusted by changing the concentration of the leveling agent in a single electroplating process and / or by changing the concentration of the leveling agent in multiple consecutive electroplating processes. It should be noted that in an exemplary embodiment comprising multiple consecutive electroplating processes, each electroplating process does not need to cover exactly the same area, thus allowing for flexible shaping of the ends of these contacts.

[0040] At symbol 250, as in Figure 1 At symbol 150, substrates 210 and 220 are joined (or mated) together so that copper contact 212 contacts copper contact 222. To form a Cu-Cu bond at the interface between these copper contacts 212 and 222, pressure and heat can be applied to maintain the contact for a period of time. If the two mating surfaces of these copper contacts 212 and 222 are clean (e.g., completely clean), random atomic movement will eventually cause atoms to traverse the interface between the two surfaces, thus contaminating the interface, and over time, the interface may completely disappear. However, if the mating interfaces of copper contacts 212 and 222 are precisely matched, this may still be a slow process, typically as... Figure 1 This is an exemplary approach because atomic motion is typically random (for example, mostly random or completely random). Therefore, according to various aspects of the invention, as discussed herein, these mating surfaces may be intentionally mismatched to enhance metal-to-metal bonding.

[0041] According to various aspects of the invention, an underfill layer can be formed between the substrates 210 and 220. For example, after attachment between the copper contacts 212 and 222, an underfill layer can be formed between and / or surrounding the copper contacts 212 and 222 (e.g., by capillary underfill, molding underfill, etc.). In another example, prior to attachment between the copper contacts 212 and 222, a pre-applied underfill layer can be formed on one or more of the substrates 210 and 220 (e.g., using non-conductive paste, NCP, etc.). For example, this underfill layer formation process can be performed before the mating of the copper contacts 212 and 222 without contaminating their mating ends.

[0042] refer to Figure 3 This figure shows a perspective view 300 of an exemplary interlocking structure and engagement method according to various viewpoints of the present invention. Figure 3 The example structures and methods, or any part thereof, can be compared with other similar structures or methods discussed herein (for example, for...). Figure 2 as well as Figures 4 to 9C They share any or all of the common characteristics.

[0043] As discussed in this paper, the ends of these interconnect structures can be shaped to allow for imperfect mating. Figure 4 The diagram shown is a perspective view of an exemplary interlocking structure and engagement method according to various viewpoints of the present invention. Figure 4 The example structures and methods, or any part thereof, can be compared with other similar structures or methods discussed herein (for example, for...). Figures 1 to 3 as well as Figures 5 to 9C They share any or all of the common characteristics.

[0044] The first (or top) copper contact 412 includes a convex surface 413, and the second (or bottom) copper contact 422 includes a concave surface 423 facing the convex surface 413. Figure 400 is a cross-sectional view taken along a longitudinal axis through the first copper contact 412 and the second copper contact 422 when they are aligned along that axis. The convex surface 413 shown in the cross-sectional view 400 contacts the concave surface 423 at the first contact point 432 and the second contact point 434. As discussed herein, this contact can extend around the convex surface 413 in three dimensions (for example, forming a circle, etc.).

[0045] The concave surface 423 includes a depth 495 between its top periphery and a bottom point (e.g., the center point of the concave surface 423). The first contact point 432 and the second contact point 434 shown in the figure are approximately located at a vertical midpoint 497 between the top periphery and the bottom point. In various exemplary embodiments, the vertical positions of the contact points 432 and 434 can reach 25% to 75% of the depth 495 from the vertical level of the top periphery into the concave region (e.g., the disc-shaped region). In various other exemplary embodiments, the vertical positions of contact points 432 and 434 may be 10% to 90% of the depth 495 by entering the concave surface from the vertical horizontal position of the periphery of the top edge, or 5% to 95% of the depth 495 by entering the concave area from the vertical horizontal position of the periphery of the top edge, or 0% to 95% of the depth 495 by entering the concave area from the vertical horizontal position of the periphery of the top edge.

[0046] For example, the first contact point 432 and the second contact point 434 may also be located between 25% and 75% of the distance from the bottom (or center) point to the periphery of the top edge in the concave surface 423. In another exemplary embodiment, for example, the first contact point 432 and the second contact point 434 may also be located between 10% and 90% of the distance from the bottom point to the periphery of the top edge in the concave surface 423, or between 5% and 95% of the distance from the bottom point to the periphery of the top edge in the concave surface 423, or greater than 5% or 10% of the distance from the bottom point to the periphery of the top edge in the concave surface 423 (for example, it may extend all the way to the periphery of the top edge).

[0047] like Figure 4 As shown, the total force F TA force F can be applied to press the first copper contact 412 and the second copper contact 422 together. This is because at contact points 432 and 434, the convex surface 413 and the concave surface 423 are not orthogonal to the applied force F. T They are in contact with each other at a certain angle in the direction of the interaction, therefore, the force F T There will be a compressive force F C This component operates to directly press the convex surface 413 and the concave surface 423 together (for example, acting orthogonally at the interfaces at contact points 432 and 434). In this text, this force F... C This force, F, can also be referred to as a positive force. T There can also be a shear force F S The component will operate tangentially at contact points 432 and 434 along the interface between the convex surface 413 and the concave surface 423.

[0048] Compressive force F C and shear force F S The bonding process operates on the metal (e.g., copper) of the first copper contact 412 and the second copper contact 422, resulting in plastic deformation. This plastic deformation may then increase the degree of atomic diffusion between the convex surface 413 and the concave surface 423. This increased diffusion degree may then lead to an overall decrease in the amount of pressure, temperature, and / or residence time required to create an effective copper-to-copper bond between the first copper contact 412 and the second copper contact 422.

[0049] Depending on the amount of deformation of the convex surface 413 and the concave surface 423, gaps may exist between individual portions of the convex surface 413 and the concave surface 423 after the metal-to-metal (e.g., copper-to-copper) bonding has been completed. It should be noted that this is not always the case. For example, during the bonding process, the convex surface 413 and / or the concave surface 423 may undergo sufficient deformation to eliminate some or all of the gaps between the convex surface 413 and the concave surface 423.

[0050] Figure 5 The diagram shown is a cross-sectional view of an exemplary interlocking structure according to various viewpoints of the present invention. Figure 5 The example structures and methods, or any part thereof, can be compared with other similar structures or methods discussed herein (for example, for...). Figures 1 to 4 as well as Figures 6 to 9C They share any or all of the common characteristics. For example, Figure 5The upper half is a cross-sectional view cut along the vertical axis (for example, the up / down direction in the example icon in this article), and for example, Figure 5 The lower half is a top view of the area near these metal-to-metal joints.

[0051] As mentioned herein, even after the metal-to-metal bonding process, gaps may remain between different individual portions of the convex surface 513 and the concave surface 523. Example drawing 500 provides an illustrative example of these gaps. For instance, the dimensions of these gaps may be enlarged for clarity.

[0052] Example 500 includes an intermediate region 562 corresponding to the plastic deformation of the convex surface 513 and / or the concave surface 523 and the metal-to-metal bonding between the first copper contact 512 and the second copper contact 522. For example, the intermediate region 562 includes regions 551 and 553, as shown in the cross-sectional view above. In one embodiment, the intermediate region 562 forms a ring surrounding the convex surface 513 and inside the concave surface 523. Additionally, for example, there is a gap in the central region 563 of the height (for example, the maximum height) D between the convex surface 513 (for example, its tip or center) and the bottom (or center) of the concave surface 523. Furthermore, for example, there is also a gap in the surrounding region 561 of the height (for example, the maximum height) P between the convex surface 513 (for example, its peripheral edge) and the peripheral edge of the concave surface 523.

[0053] However, as explained herein, while deformation of the convex surface 513 and / or the concave surface 523 may be sufficient to eliminate the gaps between these surfaces, this is not necessarily the case. For example, the convex surface 513 and the concave surface 523 may contact each other and be metal-to-metal bonded together in the central region 563 and / or the surrounding region 561. If these bonds exist, for example, their strength may be weaker than the bonds in the bonding region 562; however, this is not necessarily the case. For example, in one exemplary embodiment, the interface lines located in the central region 563 and / or the surrounding region 561 between the convex surface 513 and the concave surface 523 may be more visible (e.g., more visible) than the interface lines in the central region 562 after the bonding process.

[0054] Figure 5While example 500 is generally illustrated as having symmetrical features, this symmetry is not necessary. For example, the first contact 512 and / or the second contact 522 can be asymmetrical. This asymmetry can then lead to asymmetry in the areas where the first contact 512 and the second contact 522 are joined. For example, refer to... Figure 5 The mating area 562 may be off-center (or offset) relative to the center of the first contact 512 and / or the second contact 522 (for example, off-center by at least 5% or 10% of the contact width, or off-center by at least a certain percentage greater than the manufacturing tolerance, etc.). Additionally, for example, one side of the mating area 562 may be thinner than the other side (for example, area 551 may be shorter than area 553, or vice versa), for example, thinner than either area width by at least 5% or 10%, or thinner than at least a certain percentage greater than the manufacturing tolerance, etc. Furthermore, for example, one side of the mating area 562 may be positioned differently from the other side based on the center (for example, area 551 may be offset based on the center, while area 553 may not be offset or may be offset by a different amount). Further examples... Figure 5 The gap P shown on the left side can be different Figure 5 A similar gap on the right side (for example, at least 5% or 10% of the gap P width, or a certain percentage greater than the manufacturing tolerance, etc.).

[0055] The shapes of various convex and concave surfaces can also be described by one or more radii of curvature. Non-limiting examples of this feature are provided in... Figure 6 place, Figure 6 The diagram shown is a perspective view of an exemplary interlocking structure and engagement method according to various viewpoints of the present invention. Figure 6 The example structures and methods, or any part thereof, can be compared with other similar structures or methods discussed herein (for example, for...). Figures 1 to 5 as well as Figures 7 to 9C They share any or all of the common characteristics.

[0056] In example drawing 600, the first metal contact 612 includes a convex surface 613. For example, a portion of the convex surface 613 in a central region 655 may have a radius of curvature of R2. Additionally, for example, a portion of the convex surface 613 in intermediate regions 651 and 653 may have a radius of curvature of R3. Furthermore, for example, a portion of the convex surface 613 in surrounding regions 657 and 659 may have a radius of curvature of R4.

[0057] In addition, in the exemplary diagram 600, the second metal contact 622 includes a concave surface 623 having a radius of curvature R1. It should be noted that although the concave surface 623 shown in the figure has a single radius of curvature, this feature is provided for clear explanation. For example, multiple portions of the concave surface 623 may each include a different individual radius of curvature. In the exemplary diagram 600, R2>R1 and R3<R1. In addition, R3<R4<R1. In another exemplary implementation, R4=R3.

[0058] It should be noted that during the bonding process, due to the deformation of the convex surface 613 and / or the concave surface 623, the radius of curvature in the region where the metal-to-metal bonding is performed generally becomes the same. In a first exemplary implementation, wherein metal-to-metal bonding only occurs in the intermediate region 662 (for example, including cross-sectional regions 651 and 653), a gap still remains between the convex surface 613 and the concave surface 623 in the central region 663 and the surrounding region 661. In this exemplary implementation, due to deformation, the radii of curvature of the convex surface 613 and the concave surface 623 become substantially the same in the intermediate region 662, R2 will still be greater than R1, and R4 will still be less than R1.

[0059] In an exemplary implementation where metal-to-metal bonding eventually occurs in the central region 663 and the intermediate region 662, for example, due to deformation, the radii of curvature of the convex surface 613 and the concave surface 623 become substantially the same in the central region 663 and the intermediate region 662. In an exemplary implementation where metal-to-metal bonding eventually occurs in the surrounding region 661 and the intermediate region 662 (in addition to occurring in the central region 663, or not occurring in the central region 663), for example, due to deformation, the radii of curvature of the convex surface 613 and the concave surface 623 become substantially the same in the surrounding region 661 and the intermediate region 662.

[0060] As discussed herein (for example, with respect to Figure 5 ), Figure 6 although the example 600 of is generally illustrated as having symmetrical features, such features are not necessary. For example, the first contact 612 and / or the second contact 622 may be asymmetrical. Such asymmetry may then result in asymmetry in the region where the first contact 612 and the second contact 622 are bonded. For example, refer to Figure 6The mating region 662 may be off-center (or offset) relative to the center of the first contact 612 and / or the second contact 622 (for example, off-center by at least 5% or 10% of the contact width, or off-center by at least a certain percentage greater than the manufacturing tolerance, etc.). Additionally, for example, one side of the mating region 662 may be thinner than the other side (for example, region 651 may be shorter than region 653, or vice versa), for example, thinner than either region width by at least 5% or 10%, or thinner than at least a certain percentage greater than the manufacturing tolerance, etc. Furthermore, for example, one side of the mating region 662 may be positioned differently from the other side based on the center (for example, region 651 may be offset based on the center, while region 653 may not be offset or may be offset by a different amount). Further examples... Figure 6 R3 (and / or R4) shown on the left side may be different from R3 (and / or R4) (and / or R4) (and / or R3) (and / or R4) (and / or R3) (and / or R4) (and / or R3) (and / R4 ... Figure 6 The R3 (and / or R4) shown on the right side (for example, at least 5% or 10% of R3 (and / or R4), or a certain percentage greater than the manufacturing tolerance, etc.).

[0061] As discussed in this paper, these metal contacts can be cylindrical, rectangular, generally polygonal, etc. In various exemplary embodiments, a continuous joint line or region is not required between the convex and concave surfaces; however, in other exemplary embodiments, a continuous joint line or region may be present. Figure 7 The diagram shown is a perspective view of an exemplary interlocking structure according to various aspects of the present invention. In these examples, the metal-to-metal bonding areas are completely separated, rather than continuous.

[0062] For example, in Figure 7 In the example diagram at symbol 700, a joint with a square cross-section may be deformed and join at the four corner regions 771, 772, 773, and 774. Additionally, for example, in... Figure 7 In the example diagram at symbol 750, a contact with an octagonal cross-section may deform and engage at the eight corner regions 781 to 788. It should be noted that, as discussed herein, due to the deformation of these contacts during the bonding process, even if the initial contact between these first and second metal contacts occurs only at completely separated points, this contact may still unfold to create multiple continuously coupled engagement regions when these convex and / or concave surfaces deform, and may even unfold to create a comprehensive engagement region over the entire convex and / or concave surface.

[0063] As mentioned herein, while only a single mating interconnect pair is generally illustrated and discussed in these examples, it should be understood that many identical structures can also be formed on these substrates. For example, the contacts shown in the figures can be one of dozens or hundreds of identical or similar structures used to electrically and / or mechanically couple these substrates. For example, these contacts can be arranged in an array, a solid matrix, a square or rectangular arrangement, a straight line, multiple parallel straight lines, etc. It should be understood that while all these contacts may resemble those discussed herein, they do not need to be identical. For example, in an exemplary implementation of an electronic device having multiple such contacts, the first portion of these contacts may differ from the second portion of these contacts. Furthermore, for example, the first portion of these contacts can be as discussed herein, while the second portion can have substantially flattened ends. In addition, for example, the first portion of these contacts can be asymmetrical, while the second portion can be symmetrical. To give a further example, the first part of these contacts may have a first height, width, or spacing, while the second part of these contacts may have a second height, width, or spacing that is different from the first height, width, or spacing.

[0064] As mentioned herein, the substrate can be in various forms, including wafer form. Therefore, the scope of this invention includes wafer-level bumps (e.g., copper pillar wafer-level bumps). Examples of bump fabrication processes (e.g., wafer-level bumping processes, etc.) will be provided in... Figure 8A and Figure 9C Among them.

[0065] Figures 8A to 8E The diagram shown is a perspective view of each stage of a method for forming an interconnection structure according to various viewpoints of the present invention. Figures 8A to 8E The exemplary structures and / or methods shown, or any part thereof, can be used in conjunction with other similar structures and / or methods shown herein (for example, for...). Figures 1 to 7 as well as Figure 9C They share any or all of the common features. Although the figures show the formation of only one wafer-level bump, it should be understood that this formation process can be repeated hundreds or thousands of times on a single wafer. It should also be noted that the scope of this invention is not limited to wafer-level operations. For example, any or all of the operations disclosed herein can be performed on individual dies, die panels, or die wafers.

[0066] exist Figure 8AThe substrate 810 includes a touch pad 811 (e.g., an I / O touch pad, a bonding touch pad, etc.). It should be noted that while any number of touch pads is possible, only one touch pad is shown here for clarity. For example, the substrate 810 may include a semiconductor die (e.g., a silicon semiconductor die, an active side of the semiconductor die, a back side of the semiconductor die electrically connected to the front side, etc.).

[0067] For example, a contact pad 811 may be formed to cover the top portion of the substrate 810 at multiple contact locations. For example, the contact pad 811 may include any variety of conductive materials (for example, copper, aluminum, silver, gold, nickel, their alloys, etc.).

[0068] A dielectric layer 812 (which may also be referred to as a passivation layer) is formed on the substrate 810, for example, to cover the top side of the substrate 810. For example, the dielectric layer 812 may cover the side surface of the bonding pad 811 and / or the outer periphery of the top surface of the bonding pad 811. The dielectric layer 812 may include any variety of materials, for example, inorganic materials (e.g., silicon nitride (Si3N4), oxide (SiO2), SiON, etc.) and / or organic materials (e.g., polyimide (PI), cyclophenylbutene (BCB), polybenzoxazole fiber (PBO), bismaleimide-triazabenzene (BT), phenolic resins, epoxy resins, etc.); however, the scope of the present invention is not limited thereto. For example, the dielectric layer 812 can be formed using a wide variety of processes (e.g., spin coating, printing, spraying, sintering, thermal oxidation, physical vapor deposition (PVD), chemical vapor deposition (CVD), atomic layer deposition (ALD), etc.). For example, the dielectric layer 812 may include an aperture through which a portion of the contact pad 811 is exposed.

[0069] A UBM seed layer 813 may be formed above the dielectric layer 812 and / or above the portion of the contact pad 811 exposed through apertures in the dielectric layer 812. For example, the UBM seed layer 813 may comprise any variety of conductive materials (e.g., copper, gold, silver, metals, etc.). The UBM seed layer 813 may be formed in any variety of ways (e.g., sputtering, electrodeless plating, CVD, PVD, ALD, etc.).

[0070] like Figure 8BAs shown, a shield 821 (or template) is formed and / or patterned above the UBM seed layer 813 to define an area in which a UBM and / or interconnect structure (e.g., a metal pillar) is to be formed. For example, the shield 821 may include a photoresist (PR) material or other material, which may be patterned to cover areas outside the area in which a UBM and / or interconnect structure is to be formed.

[0071] like Figure 8C As shown, UBM 831 is formed on the UBM seed layer 813 exposed via shield 821. The UBM 831 can comprise any variety of materials (e.g., titanium, chromium, aluminum, titanium / tungsten, titanium / nickel, copper, their alloys, etc.). The UBM 831 can be formed on the UBM seed layer 813 in any variety of ways (e.g., electroplating, electrodeless electroplating, sputtering, CVD, PVD, atomic layer deposition (ALD), etc.).

[0072] Similarly, Figure 8C As shown, interconnect structure 832 is formed on UBM 831. This interconnect structure 832 can include any and all of the various features. For example, this interconnect structure 832 may share any or all of the features common to any or all interconnect structures discussed herein (e.g., contacts 212, 222, 312, 322, 412, 422, 512, 522, 612, 622, etc.). For example, this interconnect structure 832 may include copper (e.g., pure copper, copper with certain impurities, etc.), copper alloys, nickel, etc.

[0073] As discussed herein, interconnect structure 832 can be formed with a convex (or dome) end, for example, by adjusting the leveling agent concentration. For example, doubling the concentration of the leveling agent used to achieve a generally flat end can achieve a convex end (or dome) height greater than 10% or 15% of the overall height of the interconnect structure 832.

[0074] like Figure 8D As shown in the example (for instance, compared to) Figure 8C The shield 821 (e.g., photoresist) will be stripped. The shield 821 can be removed in various ways (e.g., chemical stripping, ashing, etc.). Figure 8E As shown in the example (for instance, compared to) Figure 8DThe UBM seed layer 813 (for example, at least the portion not covered by the interconnect structure 832) will be removed (for example, by chemical etching, etc.). It should be noted that during the etching of the seed layer 813, for example, at least a lateral edge portion of the UBM seed layer 813 will be etched. For example, this etching may create a cut under the interconnect structure 832 and / or UBM 831.

[0075] As discussed in this paper, an interconnect structure can be formed with a concave (or dish-shaped) end instead of a convex (or dome-shaped) end. An example of this forming process is shown in... Figures 9A to 9C Among them.

[0076] Figures 9A to 9C The figures shown are perspective views of various stages of a method for forming an interconnect structure according to various aspects of the present invention. For example, the method shown in Figure 9 may share any or all of the features with the method shown in Figure 8. Therefore, this discussion will generally focus on the differences. Figures 9A to 9C The exemplary structures and / or methods shown, or any part thereof, can be used in conjunction with other similar structures and / or methods shown herein (for example, for...). Figures 1 to 8E They share any or all of the common features. Although the figures show the formation of only one wafer-level bump, it should be understood that this formation process can be repeated hundreds or thousands of times on a single wafer. It should also be noted that the scope of this invention is not limited to wafer-level operations. For example, any or all of the operations disclosed herein can be performed on individual dies, die panels, or die wafers.

[0077] like Figure 9A As shown, as Figure 8C UBM 831 is formed on the UBM seed layer 813 exposed via shield 821. UBM 831 can comprise any variety of materials (e.g., titanium, chromium, aluminum, titanium / tungsten, titanium / nickel, copper, their alloys, etc.). UBM 831 can be formed on the UBM seed layer 813 in any variety of manner (e.g., electroplating, electrodeless electroplating, sputtering, CVD, PVD, atomic layer deposition (ALD), etc.).

[0078] Similarly, Figure 9A As shown, as Figure 8CThe interconnect structure 832 is formed on top of the UBM 831. The interconnect structure 832 can include a wide variety of features. For example, the interconnect structure 832 can share any or all of the features common to any or all interconnect structures discussed herein (e.g., contacts 212, 222, 312, 322, 412, 422, 512, 522, 612, 622, etc.). For example, the interconnect structure 832 can include copper (e.g., pure copper, copper with certain impurities, etc.), copper alloys, nickel, etc.

[0079] As discussed herein, interconnect structure 932 can be formed with a concave (or dish-shaped) end (different from the exemplary interconnect structure 832 of FIG8), for example, by adjusting the leveling agent concentration. For example, in one exemplary embodiment, a concave end can be obtained by reducing the concentration of the leveling agent used in the electroplating process. For example, a concave end (or dish-shaped) depth greater than 10% or 15% of the overall height of the interconnect structure 932 can be achieved by halving the concentration of the leveling agent used to obtain a generally flat end.

[0080] like Figure 9B As shown in the example (for instance, compared to) Figure 9A The shield 821 (e.g., photoresist) will be stripped. The shield 821 can be removed in various ways (e.g., chemical stripping, ashing, etc.). Figure 9C As shown in the example (for instance, compared to) Figure 9B The UBM seed layer 813 (for example, at least the portion not covered by the interconnect structure 832) will be removed (for example, by chemical etching, etc.). It should be noted that during the etching of the seed layer 813, for example, at least a lateral edge portion of the UBM seed layer 813 will be etched. For example, this etching may create a cut under the interconnect structure 832 and / or UBM 831.

[0081] In summary, various aspects of the present invention provide a semiconductor device for implementing metal-to-metal bonding in an electronic device and a method for manufacturing the same. For example, and without any limitation, various aspects of the present invention provide a structure and method utilizing an interlocking structure configured to enhance the metal-to-metal bonding. While specific ideas and examples have been described above, those skilled in the art will understand that various modifications and equivalents can be made without departing from the scope of the invention. Furthermore, many modifications can be made to adapt a particular case or material to the teachings of the present invention without departing from its scope. Therefore, the present invention is not intended to be limited to the specific examples disclosed; rather, the present invention covers all examples falling within the scope of the claims.

Claims

1. An electronic device, characterized in that, include: upper base plate; A metal pillar, comprising metal, the metal pillar including an upper end of the pillar coupled to the upper substrate, a lower end of the pillar, and a pillar sidewall located between the upper end of the pillar and the lower end of the pillar; The lower substrate is lower than the upper substrate in the vertical direction; A metal contact structure, comprising metal, the metal contact structure including a lower side coupled to the lower substrate and an upper side including a cavity; and A metal-to-metal solderless joint is located between the lower end of the pillar and the upper side of the metal contact structure. The metal-to-metal solderless joint is located within a first vertical height range within the cavity, but not within a second vertical height range within the cavity, wherein the first vertical height range is lower than the second vertical height range in the vertical direction. Wherein, the third vertical height range within the cavity has no metal-to-metal solderless joint, and the first vertical height range is vertically located between the second vertical height range and the third vertical height range.

2. The electronic device according to claim 1, characterized in that, The second vertical height range within the cavity includes the lateral gap between the metal of the metal column and the metal of the metal contact structure, wherein the metal of the metal contact structure surrounds the metal of the metal column.

3. The electronic device according to claim 2, characterized in that, The lateral gap does not contain conductive material.

4. The electronic device according to claim 2, characterized in that, The lateral gap is wider at the upper end of the second vertical height range than at the lower end of the second vertical height range.

5. The electronic device according to claim 1, characterized in that, The metal column has a polygonal horizontal cross-section perpendicular to the vertical direction.

6. The electronic device according to claim 1, characterized in that, Within the first vertical height range, the column sidewall is vertical, but the side of the cavity is not vertical.

7. The electronic device according to claim 1, characterized in that, In a vertical cross-section, at least a portion of the metal-to-metal solderless joint is bent.

8. The electronic device according to claim 7, characterized in that, In the vertical cross-section, the lower end of the column is bent into a profile having multiple different radii of curvature.

9. An electronic device, characterized in that, include: upper base plate; A metal pillar, comprising metal, includes an upper end coupled to the upper substrate, a lower convex end, and a sidewall located between the upper end and the lower end, wherein the lower convex end has no solder, and wherein the lower convex end includes: The first convex portion has a first convex radius of curvature; and The second convex portion has a second convex radius of curvature; The lower substrate is lower than the upper substrate in the vertical direction; A metal contact structure, comprising metal, includes a lower side coupled to the lower substrate and a concave upper side, wherein the concave upper side includes a first concave portion having a first concave radius of curvature, wherein: The radius of curvature of the first concave shape is smaller than the radius of curvature of the first convex shape; and The radius of curvature of the first concave shape is greater than the radius of curvature of the second convex shape; and Metal-to-metal solderless joint, wherein the metal at the lower end of the convex post directly mates with the metal on the upper concave side of the metal contact structure, wherein the second convex portion contacts the upper concave side at multiple points around the circumference of the upper concave side, while the first convex portion does not contact the upper concave side.

10. The electronic device according to claim 9, characterized in that, The metal in question includes copper.

11. The electronic device according to claim 9, characterized in that, In a horizontal cross-section perpendicular to the vertical direction, the metal column is polygonal.

12. The electronic device according to claim 9, characterized in that, The metal contact structure includes a metal column.

13. The electronic device according to claim 9, characterized in that, The upper substrate includes a first semiconductor die, and the lower substrate includes a second semiconductor die.

14. The electronic device according to claim 9, characterized in that, The metal column and the metal contact structure are cylindrical.

15. A method for manufacturing an electronic device, characterized in that, The method includes: Provide the first component, including: Upper substrate; and A metal pillar, comprising metal, the metal pillar including an upper end of the pillar coupled to the upper substrate, a lower end of the pillar, and a pillar sidewall located between the upper end of the pillar and the lower end of the pillar; Provide a second component, including: The lower substrate is lower than the upper substrate in the vertical direction; and A metal contact structure, comprising metal, the metal contact structure including a lower side coupled to the lower substrate and an upper side including a cavity; and A metal-to-metal solderless joint is formed between the lower end of the column and the upper side of the metal contact structure, wherein: The metal-to-metal solderless joint is located within a first vertical height range within the cavity, but not within a second vertical height range within the cavity, wherein the first vertical height range is lower than the second vertical height range in the vertical direction. The third vertical height range within the cavity has no metal-to-metal solderless joint, and the first vertical height range is vertically located between the second vertical height range and the third vertical height range.

16. The method according to claim 15, characterized in that, The second vertical height range within the cavity includes the lateral gap between the metal of the metal column and the metal of the metal contact structure, wherein the metal of the metal contact structure surrounds the metal of the metal column.

17. The method according to claim 15, characterized in that, Within the first vertical height range, the column sidewall is vertical, but the side of the cavity is not vertical.

18. The method according to claim 15, characterized in that, In a vertical cross-section, at least a portion of the metal-to-metal solderless joint is bent.

Images