Semiconductor device having a dielectric between parts
By filling the channels of polymer materials in the semiconductor device, the problem of mismatch between the thermal expansion coefficient of the semiconductor device and the supporting substrate is solved, and the reliability and functionality of the device are improved, and the risk of connection rupture is reduced.
Patent Information
- Application Number
- CN202010475307.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-05-21
- Filing Date
- 2020-05-29
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2040-05-29
AI Technical Summary
The mismatch between the thermal expansion coefficient between the semiconductor device and the support substrate leads to separation of connections, resulting in reliability issues and electrical short circuits, limiting the size and functionality of the device.
The polymer material is filled between the active portions of the semiconductor device to form a channel, absorbing stress and strain caused by thermal expansion, improving the thermal and mechanical properties of the device, so that its thermal expansion coefficient is more closely matched with the support substrate.
Improves the reliability and service life of semiconductor devices, allowing for more electrical connections and functions to be formed, and reduces the risk of connections with the substrate rupture or separation.
Smart Images

Figure CN112018054B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure is directed to a semiconductor device, and more particularly, to a semiconductor device having a plurality of active portions with polymer-filled channels therebetween. Background Art
[0002] Wafer-level chip scale packages ("WLCSP") are common in the packaging space due to their small size and efficient assembly process. A WLCSP typically includes a single semiconductor die and is coupled to a supporting substrate. One of the biggest challenges for WLCSP is the mismatch in the coefficient of thermal expansion ("CTE") between the WLCSP and the supporting substrate. This mismatch in CTE causes reliability issues because thermal cycling during operation can cause the connection between the WLCSP and the substrate to separate. The separation allows contaminants to come into contact with active areas of the WLCSP or substrate, resulting in electrical shorts and overall package failure.
[0003] More specifically, the substrate typically has a first CTE and the WLCSP typically has a different second CTE. In many cases, the CTE of the substrate is typically greater than the CTE of the WLCSP. The WLCSP is typically coupled to the substrate by solder. During operation, the resulting combination of the WLCSP and the substrate is subjected to a range of temperatures due to the heat generated by the operation. In other words, the WLCSP and the substrate are subjected to a thermal cycle from a lower first temperature when not in operation to a higher second temperature during operation before returning to a lower first temperature when operation stops. Typically, the WLCSP and the substrate are subjected to thousands of these on-off cycles and subsequent temperature cycles, if not more, during the useful life of the WLCSP.
[0004] During each of these cycles, the temperature increase during operation results in expansion of the WLCSP and substrate. However, because the CTE of each of these devices is different, the substrate and WLCSP will expand by different amounts in response to changes in temperature. Over time, this can cause the connection between the WLCSP and the substrate to break, and eventually cause the WLCSP to separate from the substrate. Once broken or separated, contaminants (such as water or dust), for example, can come into contact with the current flowing through one or both of the WLCSP and the substrate, which will cause an electrical short. The electrical short can lead to complete failure of the WLCSP or substrate, such as by overheating, catching fire, exploding, or damaging the integrated circuits or other electrical connections in the WLCSP and the substrate.
[0005] In addition, the difference in CTE between the substrate and the WLCSP creates a limitation on the size of the WLCSP, because the above problems are more complex for larger WLCSPs. In other words, using a larger WLCSP may cause problems to occur earlier than using a smaller WLCSP because there is more material to expand and there are larger and weaker connections between components. This limitation in size also limits the number of electrical connections that can be established between the WLCSP and the substrate because there is less space on the WLCSP for electrical connections. Because the limitation on the number of electrical connections limits the functions associated with the WLCSP, the limitation on the number of electrical connections limits the applications of the WLCSP. Some applications of semiconductor devices require larger semiconductor devices and more electrical connections than can be reliably produced using known WLCSPs. Summary of the invention
[0006] The present disclosure is directed to a semiconductor device having multiple active portions with channels between the portions extending at least partially through or all the way through the device. The channels are preferably filled with an elastic material, such as a polymer or other insulating and resilient buffer material. During operation, the elastic material absorbs stresses and strains within the semiconductor device due to thermal expansion of the semiconductor device. This includes stresses and strains introduced into the semiconductor via coupling with a supporting substrate (such as in a WLCSP arrangement). The ability to absorb these internal forces improves the reliability of the semiconductor device because these forces are less likely to cause the semiconductor device to crack or separate from the substrate.
[0007] The semiconductor devices described in the present disclosure also include a plurality of channels, wherein each of the channels is lateral relative to each other. In some examples, several of the channels extend all the way through the semiconductor device, while other channels extend only partially through the semiconductor device. In still other examples, all of the channels extend all the way through the semiconductor device, or all of the channels extend only partially through the semiconductor device.
[0008] A channel is formed between different semiconductor structures within a semiconductor device. For example, a channel can be formed between two different semiconductor dies that are part of a semiconductor device. Each of the dies can perform similar or different functions. Alternatively, a channel can be formed between adjacent parts or portions of the same semiconductor device. Again, the channel separates different parts or portions that perform similar or different functions. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] In order to better understand the embodiments, reference will now be made to the accompanying drawings by way of example only. In the accompanying drawings, the same reference numerals identify similar elements or actions. The sizes and relative positions of the elements in the accompanying drawings are not necessarily drawn to scale. For example, the shapes and angles of the various elements are not necessarily drawn to scale, and some of these elements may be enlarged and positioned to improve the readability of the accompanying drawings.
[0010] Figure 1 is a plan view of an exemplary embodiment of a device according to the present disclosure having a polymer-filled channel extending through a semiconductor substrate of the device.
[0011] Figure 2 yes Figure 1 2 is a cross-sectional view of the device along cut line 2 - 2 , which illustrates a channel extending through the semiconductor substrate from the first surface to the second surface.
[0012] Figure 3 is a plan view of an alternative exemplary embodiment of a device according to the present disclosure having a via extending partially through a semiconductor substrate.
[0013] Figure 4 yes Figure 3 A cross-sectional view of the device along cut line 4 - 4 is shown illustrating a channel extending partially through the semiconductor substrate and a polymer in the channel.
[0014] Figure 5-Figure 9 is a plan view of an alternative exemplary embodiment of a device according to the present disclosure having multiple channels transverse to each other.
[0015] Fig.10 yes Figure 6 1 is a cross-sectional view of a device along cut line 10 - 10 , which illustrates a first channel extending partially through the semiconductor substrate, and a second channel extending through the semiconductor substrate.
[0016] Fig.11 yes Figure 2 A cross-sectional view of an exemplary embodiment of a device coupled to a supporting substrate.
[0017] Fig.12 yes Figure 2 0026] Figure 2 is a cross-sectional view of an alternative exemplary embodiment of a similar device coupled to a supporting substrate.
[0018] Figures 13A-13H is a cross-sectional view of an exemplary embodiment of a process flow for forming a device according to the present disclosure. DETAILED DESCRIPTION
[0019] The present disclosure is generally directed to a device having a channel between active portions of the device and having an elastic material such as a polymer in the channel to increase the thermal and mechanical properties of the device. In one embodiment, the channel extends all the way through a semiconductor substrate of the device, while in other embodiments, the channel extends only partially through the semiconductor substrate. In other embodiments, multiple channels are lateral relative to each other.
[0020] The polymer increases the coefficient of thermal expansion of the semiconductor device to more closely match that of a supporting substrate, such as a printed circuit board, to which the semiconductor device is coupled during downstream processing. In other words, the material in the channel adds flexibility to the device so that expansion and contraction due to heat generated during operation is less likely to cause the connection between the device and the supporting substrate to break or separate. This improves the reliability and service life of the resulting combination. The increase in device reliability and increased flexibility enables the formation of larger devices that support additional electrical connections and functionality compared to existing devices of related types, such as WLCSP.
[0021] Figure 1-Figure 2 An exemplary embodiment of a device 100 is illustrated having a semiconductor substrate 102 having a first surface 104 opposite a second surface 106. In one embodiment, the first surface 104 is an active surface and the second surface 106 is a passive or inactive surface. The active surface includes electrical connections to active and passive circuitry formed in the semiconductor substrate 102, which may include transistors, resistors, capacitors, and other circuitry suitable for the end use of the device 100.
[0022] The channel 108 extends through the substrate 102 from the first surface 104 to the second surface 106. A material 110 is formed in the channel 108. In one embodiment, the material 110 is preferably a polymer layer, which may also be referred to as a polymer. The material 110 may also be a stress buffer or a buffer or an insulator. Alternatively, the polymer 110 may generally be any elastic material, and is preferably an elastic insulating material. In other embodiments, the material 110 may be a plastic material, a rubber material, a polyimide material (with or without graphite or glass fiber reinforcement), a dielectric material, or another type of insulating material, which preferably has elastic properties, or has a thermal expansion coefficient that is preferably greater than the thermal expansion coefficient of the substrate 102. Thus, although a variety of different materials are suitable for the material 110, the following discussion will be conducted by reference to an embodiment of the material 110 with respect to a preferred dielectric material, or more preferably a polymer material or a polymer layer.
[0023] The substrate 102 also includes a plurality of contact pads 112 on the first surface 104 of the substrate 102. Figure 2 In the embodiment, a plurality of contact pads 112 are formed on the first surface 104 and are coupled to the circuit system and metal layers (not shown) in the semiconductor substrate 102. In other embodiments, such as reference Figure 4 As depicted, the plurality of contact pads 112 may be partially recessed or formed in the first surface 104 and have an outermost surface that is in the same plane as the first surface 104. In other words, the contact pads 112 may also be between the first surface 104 and the second surface 106.
[0024] A first dielectric layer 114 is formed on the first surface 104 of the substrate 102. In one embodiment, the first dielectric layer 114 is a passivation layer. The first dielectric layer 114 may be an insulator or other dielectric suitable for the application. In one embodiment, the first dielectric layer 114 is on a portion of each of the plurality of contact pads 112 so that the surface of each contact pad 112 is exposed for coupling to other conductive layers. A second dielectric layer 116 is on the first dielectric layer 114. In one embodiment, the second dielectric layer 116 is a polyimide layer, although the second dielectric layer 116 may also be a polymer or other insulator layer. The second dielectric layer 116 is on a portion of each of the contact pads 112. An opening in the second dielectric layer 116 is formed to again expose the surface of the contact pads 112.
[0025] A conductive layer or redistribution layer 118 is on the second dielectric layer 116. More specifically, the redistribution layer 118 can include a plurality of portions, such as portions 118a, 118b formed at various locations on the second dielectric layer 116. The redistribution layer 118 is coupled to a plurality of contact pads 112. In one embodiment, each of the portions 118a, 118b of the redistribution layer 118 is in abutting physical contact with at least one contact pad 112. The redistribution layer 118 can be one of several different conductive materials, such as metals and metal alloys. As such, the redistribution layer 118 is electrically coupled to the contact pads 112.
[0026] A third dielectric layer 120 is on the second dielectric layer 116 and the redistribution layer 118. In one embodiment, the third dielectric layer 120 is a polyimide layer, while in other embodiments, the third dielectric layer 120 is a polymer or other insulator layer. The third dielectric layer 120 can be the same material as the second dielectric layer 116 or a different material. The third dielectric layer 120 includes a recess or opening 122 formed by masking, laser cutting or etching to expose a portion of the redistribution layer 118. The solder ball 124 is electrically and physically coupled to the exposed portion of the redistribution layer 118 through the recess 122.
[0027] As illustrated, device 100 includes a plurality of solder balls 124, wherein each solder ball is coupled to a respective portion 118a, 118b of redistribution layer 118, and thus, each of solder balls 124 is electrically coupled to one of contact pads 112 through redistribution layer 118. Figure 2 One solder ball 124 is shown for each portion 118a, 118b of the redistribution layer 118, but it should be understood that because the redistribution layer may extend across the first surface 104 (e.g., in directions into and out of the page in the illustrated orientation), multiple solder balls 124 may be coupled to a single portion 118a, 118b of the redistribution layer 118 to establish connections between the plurality of contact pads 112. In other embodiments, additional layers of conductive material and dielectric layers may be present to create electrical connections between the contact pads 112 on the semiconductor substrate 102 and the solder balls 124.
[0028] In one embodiment, substrate 102 is a single silicon die, wherein the silicon die includes a first logic block 126 and a second logic block 128, the second logic block 128 being separated from the first logic block 126 by a first channel 108 and a polymer layer 110. Depending on the circuit system in one or more layers of substrate 102, the first logic block 126 and the second logic block 128 can have the same or different functions. For example, the first die 126 can be a memory and the second die 128 can be a processor, although other combinations are clearly contemplated herein. The first block 126 and the second block 128 in this embodiment are physically and electrically separated by the space formed by the channel 108. The first block 126 and the second block 128 can be electrically coupled by a redistribution layer 118 and a conductive layer when appropriate to exchange signals in use.
[0029] In other embodiments, the substrate 102 includes a first die 126 and a second die 128, wherein the first channel 108 and the polymer layer 110 extend through the substrate 102 between the first die 126 and the second die 128 to separate the first die 126 and the second die 128. As described above, the first die 126 and the second die 128 can have the same or different functions and can be electrically coupled to exchange signals. In an alternative embodiment, the substrate 102 includes a first portion or section 126 separated by the channel 108, and a second portion or section 128, which can perform the same or different functions.
[0030] like Figure 1 As more clearly shown in FIG. 1 , the device 100 includes a second surface 106 that is generally flat and planar and has a polymer layer 110 that is exposed to Figure 2 In one embodiment, the polymer layer 110 includes a second outermost surface 132 (see Figure 2 ), wherein the first outermost surface 130 is coplanar or flush with the first surface 104 of the substrate 102, and the second outermost surface 132 is coplanar or flush with the second surface 106 of the substrate 102. In addition, the substrate 102 includes a first outermost edge 101 opposite to the second outermost edge 103, and a third outermost edge 105 opposite to the fourth outermost edge 107 to define the perimeter of the substrate. As illustrated, the substrate 102 is generally square, although other shapes of the substrate, including rectangular, are clearly contemplated herein. Figure 1 , the channel 108 extends between the first outermost edge 101 and the second outermost edge 103. In other words, the channel 108 extends across the entire plane of the substrate 102. However, in other embodiments, such as reference Figure 3 As depicted, the channel 108 does not extend between the edges 101 , 103 , but rather the end of the channel terminates before the edges 101 , 103 , such that a portion of the substrate 102 is between the end of the channel 108 and the outermost edges 101 , 103 .
[0031] The thickness 109 of the channel 108 is preferably between 20 microns and 100 microns, although in other embodiments the thickness 109 may be greater than or less than 20 microns, and greater than or less than 100 microns. The thickness 109 is the space formed between the first block 126 and the second block 128, such that the first block is spaced apart from the second block 128 by a dimension corresponding to the thickness. Additionally, the thickness 111 of the device 100 extending between the surface 106 and the outermost surface of the third dielectric layer 120 is preferably between 150 microns and 400 microns, or greater or less. In a preferred embodiment, the thickness 111 of the device is between 180 microns and 220 microns.
[0032] Figure 3-Figure 4 An alternative exemplary embodiment of a device 200 is illustrated, which may include the Figure 1-Figure 2 Thus, for the sake of brevity only and to avoid obscuring the features of the described embodiments, descriptions of repeated features and their functions may be omitted, but it should be understood that even without reference to Figure 3-Figure 4 For explicit description, device 200 also includes such features.
[0033] The device 200 includes a semiconductor substrate 202 having a first surface 204 opposite to a second surface 206, and a channel or recess 208 extending at least partially through the substrate 202 from the first surface 204 toward the second surface 206. In one embodiment, the first surface 204 is an active surface, while the second surface 206 is a passive or non-active surface that can be thinned to be suitable for the final use of the device. A buffer material 210 fills the channel 208, and the buffer material 210 can be a dielectric, a polymer, or a polymer layer.
[0034] The substrate 202 also includes a contact pad 212 in the first active surface 204 of the substrate (e.g., between the first surface 204 and the second surface 206). The contact pad 212 has a surface 214, which can be an outermost surface, or a surface close to the first surface 204 of the substrate that is coplanar or flush with the first surface 204. Alternatively, Figure 2 Similarly, contact pads 212 may be on first surface 204 . Figure 4 A plurality of contact pads 212 are illustrated, wherein each contact pad 212 may be in first surface 204 as described above. Alternatively, one or more contact pads 212 of plurality of contact pads 212 may be on first surface 204 while others are in first surface 204.
[0035] Device 200 is similar to device 100 (see Figure 1) is that the channel 208 does not extend all the way through the substrate 202. Instead, the channel 208 extends partially through the substrate 202 and is separated from the second surface 206 by a distance 227. As such, the substrate 202 includes a first portion 216, a second portion 218, and a third portion 220. Each of the portions 216, 218, 220 may also be referred to as a portion. As illustrated, the portions 216, 218, 220 define the sidewalls of the channel 208. In one embodiment, the thickness of the first portion 216 and the thickness of the second portion 218 are equal or substantially equal (e.g., the thickness of each portion is within ±5% of the thickness of the other portion), while the thickness of the third portion is preferably equal to or less than 50 microns (e.g., the distance 227 may be equal to or less than 50 microns). Because the channel 208 may be formed by laser cutting, etching, or other processes, the depth of the channel 208 may vary according to the specifications for the application of the device 200. Thus, in other embodiments, the thickness of the third portion 220 can be greater than or less than 50 microns, such as between 10 microns and 40 microns. As with device 100, substrate 202 can be a single silicon die, and first portion 216 is a first logic block, second portion 218 is a second logic block, with channel 208 and buffer 210 between the logic blocks. Typically, third portion 220 will not include any critical circuitry, i.e., will not include electrical connections. Any electrical connections between logic blocks or portions 216, 218 will be along first active surface 204.
[0036] like Figure 3 As more clearly shown in FIG. 1 , the second surface 206 of the substrate 202 is generally flat and planar and does not include exposed buffer material 210 because of the channel 208 shown in dashed lines, which is Figure 4 224 and preferably does not extend all the way through the substrate 202. In addition, the channel 208 preferably does not extend to the outermost edge 224 of the substrate 202, but there is a portion 222 of the substrate 202 between the end or outermost edge 226, 228 of the channel 208 and the outermost edge 224 of the substrate 202. The portion 222 is spaced a distance 225 from the edge 224 to the outermost edge 226 of the channel 208.
[0037] In a similar way to reference Figure 1-Figure 2In the depicted embodiment, the channel 208 extends through the substrate 202, but preferably does not extend to the outermost edge 224 of the substrate 202. Rather, the channel 208 terminates prior to the outermost edge of the substrate 202 such that a portion 222 of the substrate 202 exists between the end of the channel 208 and the outermost edge 224 of the substrate. In yet another embodiment, the channel 208 extends through at least a portion of the substrate 202 from the first surface 204 to the second surface 206, and extends at least partially through the substrate 202 in a different second portion of the substrate 202, rather than extending all the way through the substrate 202.
[0038] like Figure 1-Figure 4 As shown, the embodiments of the channels described herein are linear and arranged substantially parallel to at least one outermost edge of the substrate. Figure 1-Figure 4 The various channels are shown equidistant from the outermost edge of the substrate, but the channels may be spaced apart in a lateral direction (e.g., in Figure 1 Thus, the channel can be closer to one edge than the other edge in the transverse direction. In addition, in the following embodiments, such as in Figure 3 The channels in the Figure 1 In some embodiments, the channels may extend from one end of the substrate to the other end of the substrate (upward and downward in the orientation shown), it being understood that the channel may be closer to the outermost edge on one side than the other. Thus, the portion of the substrate between the ends of the channel may be different relative to each end, and more specifically, more or less at one end than at the other end. In other embodiments, one or more channels are at any lateral angle relative to the outermost edge of the substrate as opposed to being parallel. Although the sidewalls of each channel are preferably flat and planar and are configured to be at 90 degree angles relative to each other, it is also understood that etching or other forming processes may be used to form channels having sidewalls that are at lateral angles relative to other sidewalls, or include one or more rounded or curved surfaces configured such as chamfers.
[0039] Figure 5-Figure 9 Respective alternative exemplary embodiments of a device according to the present disclosure are illustrated. Figure 5-Figure 9 Each of the figures in FIG. 1 illustrates a second passive surface of a corresponding embodiment of the device. It should be understood that reference Figure 5-Figure 9 Each of the first active surfaces of the respective embodiments of the described semiconductor device may include the above reference Figure 1-Figure 4 For example, in one embodiment, reference Figure 5-Figure 9The first surface of each device described may include a passivation layer on the first surface, contacts on or in the first surface, a first polyimide layer on the first passivation layer, a redistribution layer on the first polyimide layer and coupled to the contacts, a second polyimide layer on the first polyimide layer and the redistribution layer, and solder balls coupled to the redistribution layer through openings in the second polyimide layer. As such, for the sake of brevity only and to avoid obscuring the features of the corresponding embodiments, the following description will not be repeated. Figure 5-Figure 9 Additional details of the first surface of each embodiment.
[0040] Figure 5 A device 300A is illustrated having a substrate 302A having an inactive surface 304A opposite an active surface (not shown). The substrate 302A includes a dielectric buffer 310A having a plus sign shape. A first channel region 306A and a second channel region 308A extend from the first surface through the substrate 302A to the second surface 304A to form a portion of the plus sign shape. The first channel region 306A is lateral and overlaps the second channel region 308A at a central region 316A of the substrate. The dielectric buffer 310A fills each of the channel regions 306A, 308A, and the dielectric buffer 310A may be a polymer.
[0041] In one embodiment, substrate 302A is a single silicon die that includes multiple logic blocks or portions of the die 312A that form spaces to accommodate the formation of channel regions 306A, 308A. Each logic block 312A is separated from a corresponding adjacent logic block in the plurality of logic blocks 312A by a first channel region 306A and a second channel region 308A. In other embodiments, substrate 302A includes multiple dies 312A, rather than a single die of logic blocks. In such an embodiment, each of the multiple dies 312A is separated from a corresponding adjacent die in the multiple dies 312A by a first channel region 306A and a second channel region 308A. Figure 5 As shown, channel regions 306A, 308A intersect at a 90 degree angle or substantially 90 degree angle relative to each other at a central region 316A of device 300A relative to outermost edge 314A of substrate 302A. As such, channels 306A, 308A are equally spaced from outermost edge 314A and perpendicular relative to each other.
[0042] As described above, the dimensions of each of the sections or portions 312A can be approximately 4 millimeters ("mm") by 4 mm, and each section or portion can be a separate die or a logic block of the same die. In other words, the length 318A of each side of each section 312A can be 4 mm or approximately 4 mm, wherein in this embodiment, the length of each side of each section is equal, so that the shape of each section 312A is preferably square when viewed from a plan view as shown. The overall dimensions of the device 300A can be 10 mm by 10 mm, wherein the section 312A includes 8 mm and the channel includes 2 mm of the overall dimension of 10 mm. In other embodiments, as described above with reference to Figure 1-Figure 4 As described above, the thickness of the channel regions 306A, 308A (e.g., the distance between the side walls of the channel regions 306A, 308A defined by the plurality of logic blocks 312A) is greater than or less than 2 mm, and the size of each portion 312A is greater than or less than 4 mm. The sizes of the portions 312A may also be different from each other, so that each portion has a different size relative to the other portions 312A. In other embodiments, two or more portions have the same size, while other portions have different sizes.
[0043] Figure 6 A device 300B is shown, which may include the above reference Figure 5 Some of the features described herein. The device 300B includes a semiconductor substrate 302B. The semiconductor substrate 302B includes a plurality of different portions 301B, 303B, 305B, 307B, 309B, 311B, 313B, 315B, 317B. The substrate 302B includes a first dielectric region 306B extending along a first direction and spaced apart from a first centerline of the device extending along the first direction. A second dielectric region 308B extends along a second direction transverse to the first direction and spaced apart from a second centerline extending along the second direction. A third dielectric region 310B extends along the first direction and spaced apart from the first dielectric region 306B. The device 300B also includes a fourth dielectric region 312B extending along a second direction and spaced apart from the second dielectric region 308B.
[0044] The first dielectric region 306B, the second dielectric region 308B, and the third dielectric region 310B extend all the way through the substrate 302B. As indicated by the dotted lines, the fourth dielectric region 312B only partially extends through the substrate 302B. The polymer layer 314B is in each of the dielectric regions 306B, 308B, 310B, 312B. In addition, each of the dielectric regions 306B, 308B, 310B, 312B intersects and is transverse to each other, and is equally spaced relative to the substrate 302B. In this way, there are nine parts or sections 314B in the device 300B, which can perform the same or different functions. In addition, each of the parts or sections 314B can be a logic block of a single silicon die, or can be a different die of the device 300B.
[0045] Figure 7 A device 300C is shown, which may include the above reference Figure 5-Figure 6 Device 300C includes a substrate 302C having a second surface 304C opposite to a first surface. Substrate 302B includes a channel 306C extending from the first surface to the second surface 304C. A polymer layer 308C is in the channel 306C. Figure 7 As shown, channel 306C has a diamond or square shape such that the sides of the diamond are not parallel to the sides of substrate 302C. The sides of the diamond are moved or rotated relative to the sides of substrate 302C. In some embodiments, a diagonal line starting from a corner of substrate 302C will be transverse to one or more sides of the diamond.
[0046] The described passage is not limited to linear passage, but includes other geometric shapes, which include curved shapes. In addition, passage 306C includes a first portion 310C arranged around the center 314C of device 300C, which is connected to a second portion 312C extending from the first portion 310C to the outermost edge 316C of device 300C. Device 300C also includes a first portion or portion 318C and several adjacent second portions or portions 320C, and the first portion or portion 318C is square and arranged as a rhombus, and the second portion or portion 320C has a trapezoidal shape. In this way, although some sides in the side of the second portion 320C are perpendicular to each other, other sides such as side 322C are transverse to the remaining sides. In other words, side 322C is at an angle of non-0 degree and non-90 degree relative to the remaining sides.
[0047] In one embodiment, corner 311C of the diamond may be spaced apart from the side of the die. More particularly, extension 315C extending a distance 313C from corner 311C toward the side of the die may be omitted.
[0048] Other shapes and configurations of semiconductor devices are contemplated in the present disclosure. For example, Figure 8 An alternative embodiment of a semiconductor device 300D is illustrated, with a square channel 302D and an extension 311D from a central region of the side of the square channel to the side of the device 300D. The extension 311D extends toward the central region of the side of the device 300D to provide flexibility to the device 300D. In this way, the semiconductor device 300D has a square first central portion 304D and a second portion 306D surrounded by straight lines or L-shaped. The central region 304D may include the most critical aspects of the die, i.e., those aspects that are most susceptible to stress and damage. The outer second portion 306D may include less critical circuitry that is less responsive to stress and damage.
[0049] Fig. 9 An alternative embodiment of a semiconductor device 300E is illustrated, which has a circular channel 302E, and a linear extension 311E extending toward the center area of the side of the die. Four extensions are illustrated, but two extensions may be included, such as those aligned along the center line. As such, the semiconductor device 300E includes a circular first portion or portion 304E, and a second portion 306E adjacent to and surrounding the first portion 304E on the other side of the channel 302E, the second portion 306E being generally square but having a circular edge 308E. As such, the present disclosure is not limited by the shape of the channel and the portion of the semiconductor device adjacent to the channel, but several different geometric shapes, straight shapes, and curved shapes of the channel can be expected.
[0050] Fig.10 yes Figure 61 is a cross-sectional view of a semiconductor device 300B shown in FIG. 1 along section line 10-10. As described herein, the semiconductor device 300B includes a first dielectric region 308B and a fourth dielectric region 312B. The first dielectric region 308B extends all the way through the substrate 302B from the first surface 319B of the substrate 302B to the second surface 304B of the substrate 302B. The fourth dielectric region 312B preferably extends partially through the substrate 302B. In other words, the fourth dielectric region 312B preferably extends into the first surface 319B of the substrate 302B, but preferably does not extend all the way through the substrate 302B from the first surface 319B to the second surface 304B. Instead, a portion 316B of the substrate 302B remains between an edge 318B of the fourth dielectric region 312B and the second surface 304B of the substrate 302B. In one embodiment, the thickness of portion 316B (e.g., the dimension between edge 318B of fourth channel 312B and second surface 304B) is between 10 microns and 30 microns. In other embodiments, the thickness is less than 10 microns or greater than 30 microns. In yet another embodiment, the thickness is preferably less than 50 microns. Thus, a single semiconductor device such as semiconductor device 300B can have at least one channel or dielectric region extending all the way through the substrate, and at least one channel or dielectric region extending preferably partially through the substrate, but preferably not all the way through the substrate.
[0051] Figure 11-Figure 12 The diagram shows coupling to various supporting substrates according to Figure 2 Note that various devices according to the present disclosure can be implemented in a manner similar to that described with respect to Fig.11 and Fig.12 Describe the coupling method. Fig.11 The semiconductor device 100 is illustrated coupled to a substrate 134 such as a printed circuit board ("PCB") or a flip chip PCB. In an alternative embodiment, the substrate 134 is a ball grid array ("BGA") such as a flip chip BGA, where the substrate 134 is subsequently coupled to the PCB in further processing. The substrate 134 includes a first major surface 142 opposite a second major surface 144. Preferably, the semiconductor device 100 is coupled to the first major surface 142. The solder balls 124 of the semiconductor device 100 are coupled to contacts 136 in the substrate 134. In an alternative embodiment, the contacts 136 are on the substrate 134 similar to Figure 2 1. As will be appreciated, substrate 134 preferably includes multiple layers of dielectric and conductive materials. Each layer may include a number of electrical connections, wires, or traces 138. Traces 138 are coupled to one another through dielectric layers having vias or connections 140 extending between adjacent layers.
[0052] The semiconductor device 100 is electrically connected to solder balls 148, which are coupled to contacts 146, which are on or in the second major surface 144 of the substrate 134. As such, an electrical connection between the semiconductor device 100 and an external device coupled to the substrate 134 may be established electrically by including solder balls 124 of the semiconductor device 100, contacts 136 in or on the first surface of the substrate 134, traces 138 and vias 140 of the substrate 134, and solder balls 148, which are connected to contacts 146 in or on the second surface 144 of the substrate. The solder balls 148 of the substrate 134 are electrically coupled to the external device. The electrical connections between the semiconductor device 100 and the substrate 134, including the solder balls 124 of the semiconductor substrate 100 and the contacts 136 of the substrate 134, may also be protected by an underfill material 152 formed between and around the second surface 106 of the semiconductor device 100 and the first surface 142 of the substrate 134. In one embodiment, the underfill 152 extends along the sidewalls 154 of the semiconductor device 100. Furthermore, the underfill 152 helps to physically connect the semiconductor device 100 to the substrate 134. The underfill 152 may be a molding compound or other dielectric material.
[0053] Fig.12 An alternative embodiment of a semiconductor device 100 coupled to a substrate 150 such as a PCB is illustrated, wherein instead of solder balls, wires 156 are electrically coupled between a redistribution layer 118 of the semiconductor device 100 and contacts 158 on or in the substrate 150. In one embodiment, the wires 156 are various types of metals and are soldered to the contacts 158 of the redistribution layer and the substrate such that some solder 160 remains on the contacts 158 and the redistribution layer 118. In an alternative embodiment, the wires 156 are ink printed wires in which a conductive paste having enstrained metal is deposited over a layer between the semiconductor device 100 and the substrate 150. The ink printed wires are sintered or cured, which removes the layer and hardens the wires 156. The semiconductor device 100 is coupled to the substrate 150 using a die attach material 162, which can be any known die attach material. The resulting combination is then packaged in a molding compound 164. As described with reference Fig.11 As with the described embodiments, electrical connections are established from the semiconductor device 100 through the substrate 150 so that the semiconductor device 100 can be electrically connected to external devices.
[0054] Figures 13A-13H The process flow of an exemplary embodiment of a method for forming a semiconductor device described herein according to the present disclosure is illustrated. Fig.13AThe method begins with a substrate 402, which may be, for example, a semiconductor wafer, as described herein, having a first surface 404 opposite a second surface 406, wherein the first surface 404 may be an active surface and the second surface 406 may be a passive or non-active surface. A plurality of contact pads 408 are formed on or in the first surface 404, as described herein. Then, after the contacts 408, a passivation layer 410 is formed on the first surface 404, and the passivation layer 410 may be in contact with at least a portion of each contact 408. In one embodiment, the method begins with a bare substrate 402, and the contact pads 408 and the passivation layer 410 are formed according to known processes. In an alternative embodiment, the method begins with a substrate 402 having preformed contacts 408 and a preformed passivation layer 410.
[0055] exist Fig. 13B In the embodiment, a plurality of channels 412 are formed in the first surface 404 of the substrate 402. In alternative embodiments, the channels 412 may be formed by laser cutting, plasma cutting, mechanical cutting, or etching, among other processes. Figures 13A-13H A cross-sectional view is described, but it should be understood that forming the channel 412 may include forming a Figure 5-Figure 9 402, wherein each channel is transverse to the other channels. Channels 412 preferably do not extend through substrate 402. In addition, the depth 401 of each channel 412 can vary in different embodiments, such as when it is preferred that the channels do not extend through substrate 402, in which case the channels 412 are formed at a shallower depth than in an embodiment in which channels 412 preferably extend through substrate 402. The depth of channels 412 can also vary across substrate 402, such that first channels in channels 412 are formed at a first depth while other channels are formed at a second, different depth, such as when forming Fig.10 In the illustrated semiconductor device 300B, at least one channel 412 preferably extends through the substrate 402 , and at least one channel preferably only extends partially through the substrate 402 .
[0056] exist Fig. 13C In the embodiment of the present invention, polymer 414 is deposited in each of channels 412 and cured. Polymer 414 preferably has a first or outer surface 403 that is coplanar or flush with first surface 404. Fig.13D In the embodiment of the present invention, after depositing and curing the polymer 414, the first polyimide layer 416 is formed on the passivation layer 410. Then, the redistribution layer 418 is formed on the first polyimide layer 416. The redistribution layer 418 is electrically connected to each of the contact pads 408.
[0057] exist Fig.13EIn the embodiment, the second polyimide layer 420 is formed on the first polyimide layer 416 and the redistribution layer 418. The formation of the second polyimide layer 420 includes: leaving a portion 422 of the redistribution layer 418 exposed so that Fig.13F The solder balls 424 in FIG. 4 may be coupled to the redistribution layer 418 , wherein the solder balls 424 are electrically coupled to the exposed portions 422 of the redistribution layer 418 and to the contact pads 408 through the redistribution layer 418 .
[0058] exist Figure 13G In the embodiment, backgrinding is performed on the second surface 406 of the substrate 402. In one embodiment, as shown, backgrinding includes exposing the polymer 414 in the channel 412. In other embodiments, backgrinding includes reducing the thickness of the substrate 402 (e.g., as shown in FIG. 4 ) without exposing the polymer 414. Figure 4 4. As shown, a portion of the substrate 402 remains between the second surface 406 and the channel 412. In yet other embodiments, the channel 412 is formed to different depths, such as in FIG. Fig.10 and Fig. 13B As described, back grinding includes exposing polymer 414 in some of channels 412, while at other locations where channels 412 are shallower, a portion of substrate 402 remains after back grinding. Fig.13H 4 , the substrate 402 is separated into a plurality of semiconductor devices 426, wherein each semiconductor device 426 includes at least one channel 412. In addition to etching, separating the substrate 402 into the semiconductor devices 426 may include: separation by scoring and mechanical cutting (such as using a saw), or separation by laser or plasma cutting.
[0059] Thus, an embodiment of a semiconductor device according to the present disclosure includes a semiconductor device having a channel filled with an elastic material so as to increase the CTE of the semiconductor device. The elastic material may preferably be a polymer or plastic, although other materials may also be contemplated. The polymer improves the thermal and mechanical properties of the semiconductor device and reduces the CTE mismatch between the semiconductor device and the substrate coupled thereto. In other words, the material in the channel adds flexibility to the semiconductor device so that expansion and contraction due to heat generated during operation are less likely to cause a break or separation of the connection between the semiconductor device and the substrate. This improves the reliability and service life of the resulting combination. Compared to current semiconductor devices of related types such as WLCSP, the increase in reliability and increased flexibility of the semiconductor device enables the formation of larger semiconductor devices that support additional electrical connections and functions.
[0060] In the above description, certain specific details are set forth in order to provide a thorough understanding of the various disclosed embodiments. However, one skilled in the relevant art will recognize that the embodiments may be practiced without one or more of these specific details, or with other methods, components, materials, etc. In other instances, known structures associated with semiconductor devices such as WLCSP have not been shown or described in detail to avoid unnecessarily obscuring the description of the embodiments.
[0061] Unless the context requires otherwise, throughout the following specification and claims, the word "include" and variations thereof (such as "comprises" and "comprising") are to be interpreted in an open, inclusive sense, i.e., as "including, but not limited to." Furthermore, unless the context clearly dictates otherwise, the terms "first," "second," and similar sequential indicators should be interpreted as interchangeable.
[0062] Throughout the specification, references to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least one embodiment. Thus, the phrases "in one embodiment" or "in an embodiment" appearing in various places throughout the specification do not necessarily all refer to the same embodiment. Furthermore, in one or more embodiments, the particular features, structures, or characteristics may be combined in any suitable manner.
[0063] As used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the content clearly dictates otherwise. It should also be noted that the term "or" is generally used in its broadest sense, meaning "and / or," unless the content clearly dictates otherwise.
[0064] Unless the context clearly dictates otherwise, the relative terms "about" and "approximately" when used to describe a value, amount, quantity, or dimension generally refer to a value, amount, quantity, or dimension that is within plus or minus 5% of the stated value, amount, quantity, or dimension, unless the context clearly dictates otherwise. It should also be understood that any specific dimensions of components or features provided herein are for illustrative purposes only with reference to the exemplary embodiments described herein, and therefore, unless the context clearly indicates otherwise, it is expressly intended that dimensions greater or less than the stated dimensions are included in the present disclosure.
[0065] The various embodiments described above may be combined to provide other embodiments. All U.S. patents, U.S. patent application publications, U.S. patent applications, foreign patents, foreign patent applications, and non-patent publications mentioned in this specification and / or listed in the application data sheet are incorporated herein by reference in their entirety. If it is necessary to adopt the concepts of various patents, applications, and publications to provide additional other embodiments, aspects of the embodiments may be modified.
[0066] These and other changes can be made to the embodiments in light of the above detailed description. Generally, in the following claims, the terms used should not be interpreted as limiting the claims to the specific embodiments disclosed in the specification and claims, but should be interpreted to include all possible embodiments and the full range of equivalents to which these claims are entitled. Therefore, the claims are not limited by the disclosure.
Claims
1. A semiconductor device, include: a substrate having a first surface opposite to a second surface and a first sidewall opposite to a second sidewall, the first sidewall and the second sidewall being transverse to the first surface and the second surface; a first channel extending through the substrate from the first surface to the second surface and from the first sidewall to the second sidewall; an elastic polymer in the first channel, the elastic polymer being coplanar with the first surface of the substrate and exposed to the second surface of the substrate, and the elastic polymer being coplanar with the first sidewall and the second sidewall and exposed to the first sidewall and the second sidewall; a contact pad on the first surface of the substrate; a passivation layer, the passivation layer being on the first surface of the substrate; a first polyimide layer, the first polyimide layer being on the passivation layer; as well as A redistribution layer is on the first polyimide layer, the redistribution layer being electrically coupled to the contact pad.
2. The device according to claim 1, further comprising: include: A second polyimide layer is on the first polyimide layer and the redistribution layer.
3. The device according to claim 1, in, The substrate is a single silicon die that includes a first logic block and a second logic block separated from the first logic block by the elastic polymer.
4. The device according to claim 1, in, The substrate also includes a first die and a second die, the first channel extending through the substrate between the first die and the second die.
5. The device according to claim 1, further comprising: include: A second channel extends through the substrate between the first surface and the second surface, the second channel being transverse to the first channel, the elastic polymer being in the second channel.
6. The device according to claim 5, in, The substrate is a single silicon die that includes a plurality of logic blocks, each of the plurality of logic blocks being separated from a corresponding adjacent logic block of the plurality of logic blocks by the first channel and the second channel.
7. The device according to claim 5, in, The substrate also includes a plurality of dies, each die of the plurality of dies being separated from a corresponding adjacent die of the plurality of dies by the first channel and the second channel.
8. The device according to claim 1, in, The elastic polymer includes a first outermost surface opposite to a second outermost surface, the first outermost surface of the elastic polymer is coplanar with the first surface of the substrate, and the second outermost surface of the elastic polymer is coplanar with the second surface of the substrate.
9. A semiconductor device, include: a substrate having a first surface opposite to a second surface; a channel in the first surface of the substrate, extending at least partially through the substrate and terminating at an end within the substrate before reaching the second surface of the substrate, the substrate including a portion extending from the second surface to the end of the channel; a dielectric elastic material in the channel, an outermost surface of the dielectric elastic material being coplanar with the first surface of the substrate; a contact pad in the first surface of the substrate; a passivation layer, the passivation layer being on the first surface of the substrate; a first polyimide layer, the first polyimide layer being on the passivation layer; as well as A redistribution layer is on the first polyimide layer, the redistribution layer being electrically coupled to the contact pad.
10. The device according to claim 9, further comprising: include: A second polyimide layer is on the first polyimide layer and the passivation layer.
11. The device according to claim 9, in, The substrate includes a first portion, a second portion, and a third portion, the third portion being coupled between the first portion and the second portion, each of the first portion, the second portion, and the third portion defining a sidewall of the channel.
12. The device according to claim 11, in, The thickness of the third portion is less than 50 microns.
13. The device according to claim 11, in, The first portion is a first logic block and the second portion is a second logic block, wherein the via and the dielectric elastic material are between the first logic block and the second logic block.
14. A method of forming a semiconductor device, include: forming a plurality of channels in a first surface of a substrate, the first surface being opposite to a second surface of the substrate; depositing an elastic polymer in each of the channels of the plurality of channels, the depositing comprising curing the elastic polymer; grinding the second surface of the substrate; as well as Separating the substrate into a plurality of semiconductor devices, the separation comprising: causing each of the semiconductor devices to include at least one channel of the plurality of channels; The invention also includes: before forming the plurality of channels: forming a plurality of contact pads on the first surface of the substrate; depositing a passivation layer on the first surface of the substrate; The invention also includes: after depositing the elastic polymer: forming a first polyimide layer on the passivation layer and the elastic polymer; and A redistribution layer is formed on the first polyimide layer, wherein forming the redistribution layer includes electrically coupling the redistribution layer to at least one contact pad of the plurality of contact pads.
15. The method according to claim 14, further comprising: include: After forming the redistribution layer: forming a second polyimide layer on the first polyimide layer and the redistribution layer, the forming comprising: creating a recess through the second polyimide layer to expose a portion of the redistribution layer; and Solder balls are coupled to the exposed portions of the redistribution layer.
16. The method according to claim 14, in, The grinding the second surface of the substrate includes making the elastic polymer in each channel of the plurality of channels coplanar with the second surface of the substrate.
17. The method according to claim 14, in, The grinding of the second surface of the substrate includes reducing a thickness of the substrate, a portion of the substrate being between the elastic polymer and the second surface in each channel.
18. The method according to claim 14, in, The forming the plurality of channels includes forming a first channel of the plurality of channels transverse to a second channel of the plurality of channels.
Citation Information
Patent Citations
Chip-on-wafer structures and methods for forming the same
CN103137583A
System on integrated chips and methods of forming same
CN107017171A
Semiconductor device
CN213483736U
Damascene re-distribution layer (RDL) in fan out split die application
US20160218082A1
