This is a nonprovisional application claiming the priority of Germany Patent Application No. 102017207988.6, filed on May 11, 2017, hereby incorporated by reference.
The present invention relates to a method for contacting a metallic contact pad embedded in a printed circuit board layer sequence, and to a printed circuit board comprising a metallic contact pad embedded therein.
Metallic contact pads introduced or embedded in printed circuit boards, e.g. contact pads on components such as, for example, semiconductors, power semiconductors, chips, transistors, or on ceramic layers or the like, have to be contacted for the purpose of connecting to a circuit and/or for the purpose of heat dissipation.
It is known from the prior art to expose a metallic contact pad (e.g. composed of copper) of a component embedded in a printed circuit board layer sequence and then to apply a metal for contacting the contact pad. In this case, in particular during the contacting of power semiconductors, the contact pad is often exposed over a large area in order to produce a large contact. This has the advantage of a high current conductivity and a low thermal resistance. What is often problematic in the prior art, however, is that a planar surface does not arise during large-area application of metal to the exposed region. This can lead to difficulties when the printed circuit board is subsequently applied to a heat sink.
As an alternative to the large-area exposure of the contact pads, the contact can also be produced by a plurality of parallel blind holes which can then be filled with copper galvanically. With the use of blind holes, although a planar surface is obtained, the connection is carried out only at points in a grid pattern defined by a maximum hole size and a minimum hole spacing. In this case, the maximum hole size and the minimum hole spacing are in the region of approximately 200 μm. The maximum connection of such a grid pattern is therefore less than 50%. If the hole spacing falls below the minimum hole spacing, there is the risk of the prepreg layer losing adhesion and thus of a reliable metallization no longer being able to be produced.
Taking this as a departure point, a method and a printed circuit board having the features as disclosed herein are proposed. Embodiments of the invention are described herein.
According to the invention, for the purpose of contacting a metallic surface of a contact pad embedded into a printed circuit board layer structure, firstly a number of first holes are introduced into a surface of the printed circuit board configuration and said number of first holes are subsequently filled with conductive material. A number of second holes are subsequently introduced between the filled first holes into the surface of the printed circuit board configuration in order likewise subsequently to be filled with conductive material.
The invention is based on the insight that the production of holes for exposing the contact pads is advantageous by comparison with large-area exposure of the contact pads. The dimensions of the holes can be chosen such that the subsequently applied metal layer has an optimum surface and in particular a good planarity. This facilitates further process steps such as, for example, applying the printed circuit board to a heat sink, since, on account of the high planarity of the metal surface, either the latter can be applied directly to the heat sink or subsequent surface processing steps (e.g. grinding) are at least less complicated.
According to the invention, the method comprises the following steps:
The plurality of first holes introduced initially together form a first hole matrix. In the context of the present invention, the term “hole matrix” denotes a totality of the holes situated in a specific arrangement with respect to one another.
In a further step, the first holes are at least partly filled or filled in by the application of a metal layer. The exposed regions of the metallic contact pad thereby come into contact with the applied metal layer.
In addition, a plurality of second holes are produced in the surface of the printed circuit board layer sequence and the metallic contact pad is at least partly exposed in this way. The plurality of the second holes produced in this step together form the second hole matrix. The holes of the second hole matrix are offset relative to the holes of the first hole matrix. This means that the holes of different hole matrices are arranged at different positions. The positions of the holes of the second hole matrix can be chosen such that they do not overlap the holes of the first hole matrix, or they can also be chosen in an overlapping fashion (an overlapping arrangement can serve, if appropriate, to remove dielectric residues from the corners of the first holes). The holes thus produced are then likewise at least partly filled by a metal layer.
By way of example, the production of holes is carried out by means of a process sequence comprising removing the copper by means of etching technology, followed by eroding the dielectric as far as the metallic surface to be linked, such as by means of a laser ablation process that acts selectively on the dieletric.
In known production methods, the holes may be able to be produced only with a specific minimum spacing with respect to one another. By way of example, with known laser drilling methods within one process step it is possible to produce holes having a diameter of a maximum of approximately 200 μm with a minimum hole spacing of approximately 400 μm. This inevitably gives rise to an interspace between the holes, which interspace cannot be utilized for the contacting. The area utilization of a contact produced in this way is only approximately 35% in this case. If the hole diameters are chosen to be larger, the process of filling with Cu electrolytically/galvanically does not function any more since the metallization then traces the profile, which leads to a deeper metallized area with the disadvantages described. If the hole spacing is chosen to be significantly smaller, there is the risk of the dielectric losing adhesion on the metallic contact and the subsequent electrolytic processing being carried out erroneously.
By providing the second hole matrix according to the invention, the holes of which are offset relative to those of the first hole matrix, it is possible, however, for the contact pad area present to be utilized significantly better. In this regard, a large contact pad having a higher thermal mass and thus a low thermal resistance and a high current conductivity and moreover a lower on resistance is produced, at the same time a good planarity of the contact surfaces being maintained or obtained (improved linking). The hole matrices can be produced for example by laser drilling or by an etching method in each case followed by dielectric erosion by means of laser processing.
The first hole matrix and the second hole matrix can be formed respectively by a first grid and a second grid. In this context, the term “grid” means that the respective holes have a regular arrangement relative to one another. In this case, the holes have identical spacings from one another along a spatial direction. The regularity can exist both along a first spatial direction and along a second spatial direction different than the first spatial direction. In this case, the holes have identical spacings from one another along the second direction as well. The spacings of two holes along the first spatial direction can correspond to the spacings of the holes along the second spatial direction.
The first and/or the second grid can additionally be rectangular. This means that the first spatial direction is oriented perpendicularly to the second spatial direction. This facilitates the production of the holes since production methods for producing holes are often designed for a rectangular arrangement.
The arrangement of the holes in a grid can be described for example with the aid of an imaginary coordinate system and a two-dimensional basis {a,b} consisting of the basis vectors a and b that span the grid. In this case, the basis vectors are different than zero and point in different spatial directions. A hole matrix forms a grid if the position of each hole proceeding from an origin of the imaginary coordinate system can be described by an addition of the basis vectors in accordance with (n·a+m·b) wherein n and m are to be selected from the set of integers. The position of a hole is then unambiguously defined by the indication of the values (n,m). In this case, the origin of the imaginary coordinate system can be placed for example at the midpoint of one of the holes. A hole is then situated at the origin of the imaginary coordinate system, and also for example at the positions (1,0), (2,0), (3,0), (0,1), (0,2), (0,3), (1,1), (1,2), (2,1), etc. Holes arranged in a grid can be produced significantly more simply. Moreover, on account of the regular arrangement of the holes in a grid over the entire contact pad area the contact pads are contacted homogeneously, which is accompanied by homogeneous contact properties.
Provision can be made for the first grid and the second grid to be spanned by a two-dimensional basis {a,b}, wherein the holes of the first grid are arranged at the locations (n,m), and wherein the holes of the second grid are arranged at the locations ((n+½),(m+½)), wherein n and m are to be selected from the set of integers. In this case, the first grid is offset relative to the second grid such that each hole of one grid is situated in the middle between two holes of the respective other grid. This increases the homogeneity of the contact even further and additionally leads to good space utilization.
The holes can have an arbitrary shape, in principle. In possible embodiments, the hole matrices comprise circular holes and/or rectangular holes. By way of example, the holes can also have a square shape. In the case of square holes, the first hole matrix and the second hole matrix together can form a checkered pattern. As a result, virtually the entire contact pad area available can be utilized for contacting.
A further embodiment of the invention provides for a third hole matrix having a plurality of third holes to be produced in the surface of the printed circuit board layer sequence in order to partly expose the metallic contact pad, wherein the holes of the third hole matrix are arranged in a manner offset relative to the holes of the first and second hole matrices. After introducing the third hole matrix, applying a metal layer is in turn provided, such that the holes of the third hole matrix are at least partly filled with metal. The offset arrangement means here that the holes of the third hole matrix are arranged at different positions than the holes both of the first hole matrix and of the second hole matrix (in an overlapping or non-overlapping fashion, as described above). With the use of a further hole matrix, it is possible, if appropriate, to utilize an even greater proportion of the contact pad area for contacting.
The invention can be developed in an analogous manner by introducing further hole matrices with subsequent metal coating in each case.
By way of example, an electrolytic deposition method known to the person skilled in the art is appropriate for applying the metal layer, said deposition method depositing higher layer thicknesses in holes than on the surface. In this case, the process can be implemented such that an excess elevation of the coating metal arises above the holes to be filled, that is to say that the hole is as it were overfilled. This process implementation has the advantage that these excess elevations can be ground away in a subsequent work process, which contributes to a further improvement in the planarity of the surfaces.
The invention additionally relates to a printed circuit board comprising a printed circuit board layer sequence and a component embedded therein, said component having at least one metallic contact pad, wherein the at least one metallic contact pad is contacted with the aid of the method according to the invention. The printed circuit board can be developed by further features described in association with the method according to the invention.
Further advantages and configurations of the invention are evident from the description and the accompanying drawing.
It goes without saying that the features mentioned above and those yet to be explained below can be used not only in the combination respectively indicated, but also in other combinations or by themselves, without departing from the scope of the present invention.
The invention is illustrated schematically on the basis of exemplary embodiments in the drawing and is described thoroughly below with reference to the drawing.
A component 30 is embedded into the printed circuit board layer sequence 10. In the exemplary embodiment illustrated, the layer sequence comprises a substrate layer 12 and a copper inner ply 14 applied thereon, the component 30 being applied on said copper inner ply by means of a contact layer or connection layer 16 (sintering layer or solder layer). The substrate layer 12 can be a conductive or a non-conductive printed circuit board material, such as an FR-4 inner ply, for example. In the former case, the copper inner ply 14 can be dispensed with.
The component 30 is embedded in a prepreg or dielectric layer 18, which terminates with a copper foil 20 applied thereon. The latter could also be omitted in one variant, such that only a dielectric layer is used. In this case, the surface and the hole walls would have to be coated conductively and with adhesive strength before copper can be deposited. That could be effected for example by sputtering of copper or chemical deposition.
The printed circuit board structure illustrated and described is purely by way of example, and any other structure deviating therefrom is readily apparent to the person skilled in the art.
The component 30 can be—as in the exemplary embodiment illustrated—a field effect transistor, comprising a semiconductor body 32 and terminals in the form of contact pads arranged thereon, which are designated by the reference signs 34 for the gate contact pad, 36 for the source contact pad, and 38 for the drain contact pad.
It is readily apparent to the person skilled in the art that any other form of component with corresponding contact pads for contacting can be used according to the invention.
The method according to the invention is explained below with reference to
Firstly, e.g. by means of a laser drilling method or etching of the copper foil and erosion of the dielectric using a laser, substantially circular holes L1 are produced in the surface of the printed circuit board layer sequence 10. The copper foil 20 and the underlying dielectric 18 are removed as a result and the contact pad 36 is partly exposed. In the example illustrated, the holes L1 have a diameter of 100 to 200 μm. The holes have a spacing of 500 μm with respect to one another both along a first spatial direction and along a second spatial direction. In this case, the first spatial direction is perpendicular to the second spatial direction, such that the totality of the holes L1 forms a first square grid. This will be explained in greater detail below with reference to
Afterward, a metal layer 11 (also extends thinly on the copper foil 20) is applied, e.g. with the aid of an electrolytic deposition method or some other method known per se to the person skilled in the art, in order to fill the holes L1 with metal. This is illustrated in
After the filling of the holes L1 of the first hole matrix, a second grid is produced in a subsequent process step, said second grid being formed by a plurality of holes L2. The second grid is likewise square with a hole spacing of 500 μm and a hole diameter of 200 μm. The second grid is offset relative to the first grid, such that the holes L2 of the second grid are situated in each case in the middle between two holes L1 of the first grid (and vice versa).
The state after producing the holes L2 of the second grid is shown in
Finally, in a subsequent process step, e.g. by means of the electrolytic filling method already mentioned above, a metal layer 13 is once again applied in order to fill the holes L2 of the second grid (cf.
In order to clarify the mathematical description of the grids already used above, an origin 21 of an imaginary coordinate system and the basis vectors a and b are additionally depicted in
With use of a single hole matrix for contacting a contact pad, as known from the prior art, given a hole diameter of 200 μm and a hole spacing of 500 μm only approximately 35% of the existing area is utilized. By virtue of the second hole matrix according to the invention, the area utilization is increased to as much as 75%.
According to the invention—as shown e.g. in the illustration in
Afterward, the holes—identified by “L2”—of a second hole matrix were produced and filled with metal. Finally, the holes—identified by “L3”—of a third hole matrix were produced and filled with metal. In terms of their geometric dimensions and in terms of their relative arrangement the three hole matrices are fashioned such that together they form a hexagonal close-packed structure. This means that each hole of a hole matrix is surrounded by six adjacent holes of the other two hole matrices arranged at identical distances. The area utilization can be increased even further as a result.
By gradation of the square sizes, e.g. 200 μm edge length in the first pass and 220 μm edge length in the second pass, it is possible to produce overlap regions which compensate for any possibly occurring offset and, as a result of the overlap, would lead to the dielectric being removed without residues. The linking/connecting to the contact pad would thus be 100% in conjunction with a leveled surface.
Number | Date | Country | Kind |
---|---|---|---|
10 2017 207988.6 | May 2017 | DE | national |