Here is the abstract you requested from the IMAPS_2016 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Electrical Characterization of Low-Profile Copper Foil for Reduced Surface Roughness Loss|
|Keywords: Material Characterization, Surface Roughness, Profile|
|Current computer systems require communication between integrated circuits through electronics packaging at increasingly higher clock frequencies. A major issue for reliable transmission of such high-speed digital signals is the loss in transmission lines. The surface roughness of copper foil has a direct impact on high-frequency loss. As signal frequencies approach 10 GHz and beyond, the variation of the conductor loss with frequency depends on the root mean square (rms) of surface roughness. The classical skin-effect resistance with a square-root-of-frequency dependence does not hold as the skin depth becomes smaller than the surface roughness. Hence, knowledge of the rms value of the surface roughness is needed for accurate modeling of loss in transmission lines. In this paper, we demonstrate a methodology for extracting the surface roughness parameters through high-frequency measurements. We use a shorted cavity resonator for material characterization, as it eliminates radiation and is completely filled with the material to be tested. The cavity resonators include different surface treatment process of electrodeposited and low-profile copper. The low-profile copper has smoother surface finish, hence reduces the loss in transmission lines. We characterize the core layers with the different surface treatments as well as the prepregs on the same board. The general procedure for the presented material extraction methodology is based on fitting simulations to measurements. The measurements are taken by using a vector network analyzer and impedance parameters are obtained. Then full-wave simulations are run by varying the surface roughness parameters as well as the dielectric constant and loss tangent. By comparing the measurement and the simulation results, we can extract a set of parameters when the simulations and measurements match well. For accuracy of the geometrical parameters used in the simulation, cross sections are taken for the cavities. The dielectric thickness as well as the surface roughness profiles are measured. The rms value of the surface roughness is calculated from the cross section using a mathematic model. This is then compared with the extracted surface roughness through high-frequency measurements. The major contributions of this paper can be summarized as follows. Even though it is well known that the surface roughness rms value is a critical electrical parameter, there has been no specific method for calculating it from the cross section picture. In this paper, we present clearly how to calculate the rms value of roughness from the geometry. Also, we correlate the geometrically extracted surface roughness value with its high-frequency characterization. Therefore, we intend to close a critical gap between geometrical models and electrical performance in surface-roughness loss characterization.|
San Diego State University
San Diego, California