Patent Application
20240011181
The field of the invention relates to the use of a mechanical bond when bonding plastic to metals in integrated circuit components.
Plastics used in semiconductor manufacturing and packaging have relatively low surface energy compared to the other materials the plastics are to interact with as part of a chip. This comparatively low surface energy makes it hard to directly bond materials, including, for example, copper or nickel-iron, to the surface of the plastic.
One potential way to increase the surface energy available from plastic is to increase its surface area. The surface area of the plastic can be increased with a mechanical pretreatment of the plastic, such as sanding, sandblasting, or grinding the plastic. A mechanical pretreatment generates a roughened surface on the plastic that will allow for increased surface energy to become available to the system as well as a mechanical bond between the plastic and whatever is plated onto its surface. The mechanical bond occurs because the pretreatment will generate a rough surface with peaks, valleys, and overhangs in the plastic surface that will cause the plastic to physically interlock with material plated onto it. This interlocking helps to secure the plastic and plated material together. The mechanical bond is in addition to any chemical bond between the two materials. Thus, the mechanical pretreatment creates two major benefits 1) the increased surface energy and 2) the mechanical bonds from the roughened plastic surfaces.
However, as noted above, these benefits rely on a mechanical pretreatment that roughens the plastic surface. This becomes an issue when the material plated onto plastic is intended to serve as a pathway, especially for current and magnetic flux; for example, a metal plated onto a roughened plastic surface would follow the contours of the plastic surface and present a longer current or flux path than a straight line of metal would. This extended pathway increases the system's resistivity and thus reduces its performance. Therefore, to allow for the mechanical pretreatment of plastic to occur without affecting the performance of bonded metals, the roughness of the metal components must be reduced.
Therefore, a solution is needed to utilize mechanically pretreated plastic without performance loss due to the increase in pathway length caused by the associated increase in roughness.
The present invention reduces or eliminates the negative effects of forming conductive or magnetic layers on base layers of, for example, epoxy plastics, in relation to surface roughness—allowing the surface of a base layer to even be intentionally roughened for increased mechanical bonding and available surface energy without a significant reduction in component performance.
These benefits are achieved by lowering the average roughness of the conductive or magnetic component through the utilization of component shapes with a method that is relatively low-cost to implement. The result is a component material layer stack, comprising, in at least one exemplary embodiment of at least one material with a U-shape cross-section, a portion of an outer surface of the U-shape material having a rough contour, and the rough contour surface of the u-shape material operably bonded to a roughened portion of a base layer. The u-shaped component can be encapsulated in semiconductor packaging to form a semiconductor packaging layer which can then be easily integrated into multi-layer packaging.
It will be appreciated that the outer surface portion of the U-shape, which has a rough contour, is typically the foundation of the U-shape, and it achieves its rough contour by following the contours of a roughened base onto which it is deposited and bound. However, the vertical sides of the U-shape, both the exterior and interior portions, do not typically follow a roughened contoured shape but take up the smoothness of the patterning material used to form the layers. These smooth and straight vertical sides reduce the average roughness of the u-shape component allowing for more efficient performance when compared to a component strictly following the contours of the roughened base. The U-shape provides greater surface area than a solid block shape and, therefore, has a stronger reduction in the average surface roughness of the component than a block of material.
The terms rough and smooth refer to the preparation of a surface, a surface that has been treated in some manner to produce a surface with less roughness than the initially occurring surface is smooth, while the unsmoothed surface is rough. For example, a semiconductor plastic base layer will initially have a surface roughness limited by the means of forming the layer, but the surface roughness may then be decreased by a smoothing process, for example, grinding or polishing. In a similar vein, when a surface has been intentionally roughened it may be referred to as a roughened surface. A rough contour is formed when a portion of the material is deposited along a roughened surface.
Therefore, by providing a means to reduce the average roughness of a component layer, the negative effect of roughened base, which is to create inefficient rough contours within material deposited onto it, is reduced. This reduction allows for components of similar efficiencies to their now roughened-contoured counterparts now enjoy the benefits of increased mechanical bonding as well as greater available surface energy.
It will be appreciated that the formation of the layer stack of the present invention may be achieved by: roughening an initial base layer; depositing a seed layer onto the initial base layer; creating a pattern for a series of pillars on the seed layer, the series of pillars having a first pillar and a last pillar; depositing pillars according to the pattern; removing the pattern; covering the first pillar and the last pillar with a protective covering while leaving the remaining pillars exposed; plating at least one layer of material onto the exposed pillars; covering all pillars in a protective covering so that the height of the covering exceeds the height of the pillars, creating a structure; grinding the structure down until an upper surface of each of the pillars is exposed, covering the exposed surface of the first pillar and the last pillar in the series of pillars with a protective covering while leaving the remaining pillars exposed, etching the exposed pillar and seed layer; removing the protective covering; and adding a build-up film.
In at least one exemplary embodiment, the result is a U-shaped material with an average surface roughness (Ra) of less than five microns. When roughening a semiconductor plastic base by sanding, surface peaks reach their highest at five microns, and therefore a component with a surface roughness of less than five microns is desirable.
As noted above, the U-shaped material may be a magnetic core component. However, it will be understood that core components may travel along a plane of the layer, and therefore the cross-section, not the full component shape, is u-shaped. A magnetic core component may be made of many materials, for example, but not limited to, nickel-iron, cobalt, or other ferromagnetic materials, including ferromagnetic materials and core compositions with non-magnetic additives.
These material component layers may be stacked to achieve multi-layer stacks of u-shaped component materials. The formation of such a stack may be achieved by further comprising repeating the steps of the initial layer stack formation, from depositing a seed layer onto the base layer through the adding of the build-up film, to form at least one additional layer on the prior formed layer with each newly formed layer now acting as the base layer for each directly subsequent layer.
It will be appreciated that in such cases where layers of core material are stacked, it is beneficial to reduce eddy current and fringe effects by offsetting the magnetic core material of each layer, and therefore, in the present invention, layers with core or conductive material may be offset.
It will also be appreciated to further reduce the potential interaction between core or conductive materials of each layer the layers may be separated. In at least one exemplary embodiment of the present invention, the layers are separated by a pillar extension that extends from the top of a base layer and upon which the subsequent layer is built. The layer extension allows for gaps within the layers, keeping the layers separated. These layer extensions may be formed on a component layer of the present invention by grinding down the build-up film to expose the first and last pillars, plating onto the first and last pillars a layer extension, and repeating the steps of depositing a seed layer onto the base layer through the adding of the build-up film, to form an additional layer, where the initial base is replaced with the layer extension, wherein the first and last pillars of each adjacent layer are connected by the layer extension and each layer, but for the layer extension, is spaced from the next layer.
Further other forms of separating layers may be included, for example, separating the layers of U-shaped components of the present invention with seed layers or other additional component layers as suitable for the purpose of the overall component.
Therefore, the u-shapes materials and the resulting layer stack of the present invention are suitable for multi-layer packaging and can serve to reduce the negative effects of utilizing a roughened base layer preserving the intended performance of a component, for example, an inductor, having a core of u-shaped materials.
A system is provided that reduces the average roughness without increasing the required pathway length-resulting in layers that can achieve the benefit of mechanical bonding without significant sacrifice to performance loss. For example, in relation to a magnetic core, a series of U-shaped nickel-iron components are plated onto a semiconductor plastic in a manner that allows the U-shaped nickel-iron components to have a low average roughness. The lowered average roughness reduces the path length the U-shaped nickel-iron components provide for magnetic flux. The vertical surfaces of the U-shaped have a surface roughness comparable to dry film, while only the bottom surface of the U-shaped nickel-iron components presents a surface roughness matching that of the rough plastic they are plated on.
The result is shown in
The U-shape components have two vertical portions 101 and a horizontal portion 102. The horizontal portion will typically contain the rough contour as shown in
A copper seed layer is then deposited onto the roughened surface of the plastic by electroless plating. The copper seed layer will serve as a seed layer that follows the contours of the plastic surface.
A dry film is placed onto the copper seed layer, and a series of copper pillars are patterned. The dry film process may be a negative or a positive process. In either process, the dry film is etched by light, presenting a smooth vertical surface in the patterning. The dry film patterning should provide at least two copper connector pillars for multi-layer packaging.
Copper pillars are then plated into the dry film pattern. The copper pillars' vertical surfaces will have a much lower average roughness than the horizontal surfaces. The bottom horizontal surface of the copper pillar will have an average surface roughness of the plastic underneath it. At the same time, the top horizontal surface will have a surface roughness not controlled by the dry film surfaces or plastic and thus be slightly rougher than the vertical surfaces but still far less rough than the bottom horizontal surfaces.
After the copper pillars are plated, the dry film overall non-connector pillars is removed, and nickel-iron is plated. The nickel-iron will follow the contours of the copper and take up the surface roughness of the copper beneath it. Therefore, the vertical surfaces of the nickel-iron will be extremely smooth compared to the plastic surface. The bottom surfaces of the nickel-iron will take the path of the plastic; however, due to the smoother vertical surfaces, the average roughness of the nickel-iron components is reduced.
Once the nickel iron is plated, a dry film is placed to cover all the nickel iron. The dry film can be vacuumed-sealed or rolled into the gaps between the pillars. Once the dry film is placed, a grinding process occurs so that the entire structure is ground down to a level exposing the upper horizontal surfaces of the copper pillars under the nickel-iron. This grinding process leaves the nickel-iron in a series of repeating U-shapes by exposing the upper surface of the copper pillars.
The copper pillars, but for the connector pillars, are then etched out, and any remaining dry film is removed. Ajinomoto Black Build-up Film is then added to complete the layer.
It will be appreciated that further layers can be made using the same process but built upon the initial layer instead of the roughened plastic. When the layers are built directly on each other, the U-shaped nickel-iron components are to be offset in every layer to avoid creating nickel-iron loops, as nickel-iron loops generate strong eddy currents, which would reduce the performance of the system. These layers may also include other components or packaging layers between each layer as is suited or dictated by the best practices of forming the overall component (for example, an integrated inductor).
Further layers can also be created with a gap in-between the layers by extending the copper connector pillars above the initial layer and then using the extension to connect the layer. When this occurs, the U-shaped nickel-iron components need not be offset because the gap prevents a loop from forming. However, this takes up more space than stacking the layers directly on each other.
Now, more specifically, we shall approach the steps of the present invention in greater detail.
When the base layer 110 is ready to receive the component layer, here referring to a layer of U-shaped component material, a pattern of pillars is to be produced.
One material that is useful for forming the pillars is copper, which enables strong electrical connections. However, many different pillar materials may be used, especially for those pillars, which will only be utilized for structural support during the component layer formation to reduce costs.
Once the pillars have been formed, the dry film is removed, as shown in
When the material which will form the U-shaped components is placed it will follow the surface contours of the stack 800 and take up the smoothness of the vertical portions and the roughness of the base for the horizontal portions. Before placing the material, which will become the UI-shaped components it is good to cover the pillars, which will become the electrical connections to other layers, with a dry film or other protective covering to prevent the material from plating onto the electrical connection pillars. This can be seen in
As shown in
This grinding step has also separated the material 1001 into a series of U-Shapes 100. It can be seen that there are two L-shapes 105 of material as well. These L-shapes 155 were formed by the dry film 150 used to protect the first 151 and last pillars 152. By patterning a variety of shapes, many forms of material can be produced.
Once the grinding step is completed, the pillars are removed. In at least one exemplary embodiment, when copper pillars have been used, the pillars are removed by etching. As shown in
The result is a U-shaped component which, although it does absorb the base layer roughness, has its average roughness reduced by the vertical portions 101 of the U-shape.
When a multi-layer package is desired, the upper surface of stack 180 may be ground down to expose the first pillar 131 and the last pillar 132, as shown in
Additional layers containing U-shape components are shown in
As noted above, additional layers may be interspersed with the U-shaped component layers. One such layer stack 1500 is shown in
To further reduce parasitic eddy currents, when utilizing additional layers with u-shaped components, a space can be created between the layers. As shown in
As demonstrated above, the result of the present invention is a layer stack that operates efficiently even though it has a roughened base as the initial U-shape component layer absorbs the roughness. The roughness, in turn, allows for greater physical and mechanical bonding to occur, strengthening the layer stack significantly.
The drawings and figures show multiple embodiments and are intended to be descriptive of particular embodiments but not limited with regard to the scope or number, or style of the embodiments of the invention. The invention may incorporate a myriad of styles and particular embodiments. All figures are prototypes and rough drawings: the final products may be more refined by one of skill in the art. Nothing should be construed as critical or essential unless explicitly described as such. Also, the articles “a” and “an” may be understood as “one or more.” Where only one item is intended, the term “one” or other similar language is used. Also, the terms “has,” “have,” “having,” or the like are intended to be open-ended terms.
| Number | Date | Country | |
|---|---|---|---|
| 63359918 | Jul 2022 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 18131350 | Apr 2023 | US |
| Child | 18220756 | US | |
| Parent | 18135660 | Apr 2023 | US |
| Child | 18131350 | US |
