Here is the abstract you requested from the imaps_2018 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.
|Electrochemical capacitance based method applied to epoxy molded devices|
|Keywords: System In Package, epoxy morphological irregularity, electrochemical measurement method|
|The most advanced microelectronics systems are manufactured by connecting different chips types to combine the specific advantages offered by each material application. A typical example of such complex systems is the pairing of a micro-mechanical structure (MEMS) with an electronic Application Specific Integrated Circuit (ASIC) chip with the aim to detect movements, pressure differences, position etc. These System in Package (SiP) systems are commonly packaged with an epoxy resin material because of many advantages such as lower cost, lighter weight, and better performance over hermetic packages, despite the moisture absorption after the exposure to the humidity. The wires connecting the devices consequently are in contact with a compound that is not a waterproof material. It is well known that moisture diffusion in epoxy molding compounds (EMCs) is one of the major reliability concerns in plastic encapsulated microcircuits (PEMs), because many failure modes observed in these devices are believed to arise from the diffusion of moisture during manufacturing, storage, or operation. In fact during the plastic package molding process, the resin components distribution (polymer and silica fillers) around the semiconductor system devices is influenced by the chips geometry profiles (and their technology) and the epoxy physical/electrical property may have a certain distribution throughout the molding compound. This article will describe the real resin morphology near the typical polysilicon surfaces of a MEMS device in the area of the pads carrying the electric signals to be detected by a dedicated IC. The epoxy structure close to the above regions may be affected by a non-homogeneous distribution of the molding components that can locally increase the compound humidity sensitivity. Also the electrical properties are obviously conditioned by this morphological irregularity. The goal of this study is the identification of an appropriate measurement method to characterize the local deviation of the resin electrical properties from the standard expected behavior depending on the environment condition such as temperature and humidity. Certainly a typical concentrated parameters model as an R-C-R network is the simplest way to correctly address the resin behavior study. The measurement equipment to be used and the setups are connected to the target electrical parameters, the scope and the frequency of the involved phenomenon. The parameters magnitude is intrinsically low because of the limited volume of the material under investigation. In case of the resistance parameter evaluation, a typical commercial current amplifier may be used with a current resolution of the order of few pA according to the standard specification of the epoxy compound. The capacitance parameter measurement needs an instruments setup more sophisticated than that required by a pure DC analysis. A trans-admittance network analyzer is a good preliminary choice to qualitative list the electrical parameters of the network locally activated by the compound irregularities. The output signals give the profile of a transfer function whose zero and pole are the base of the network study. However any electrical investigation that does not take into account the type of the physical and the chemical effect of the material modification after moisture absorption may lead to a wrong modelling of the phenomenon. A deep knowledge of the resin structure especially in the complex geometry around the sensor device pads is necessary to pave the way for a suitable measurement method. The morphological analysis of the SiP epoxy molded devices is the right structural work to be completed before to select the appropriate electrical study approach. The physical aspects of the epoxy compound will be extensively detailed after opening some parts by selectively removing the resin. The X-Ray computed tomography of the most critical areas is clearly showing the presence of filler voids near the MEMS pads. As the people skilled in the art well know the lack of such waterproof material inside the resin increases the humidity sensitivity. A lot of irregularities examples will be reported as due to the injection condition of the molding material during the production phase. The fillers having a size comparable to the pads to pads distance make a kind of a shadow masking effect against the small fillers injection so to finally generate visible filler voids in the most critical areas. A first measurements setup will be described by probing the wires connected to the pads and by analyzing the results with a traditional electronic method. The sample under investigation will be submitted to a thermal humidity stress (THS) cycle after a baking session in order to start the experiment from a dry condition. The measure results of the dry and the damp session will be compared in order to capture a potential electrical parameters difference. This pure electronic equipment setup is not able to give the results to build the equivalent circuit modelling the resin local structural differences. The chemical aspects of the phenomena activated in the resin after the moisture exposure must be considered to identify a methodology that better fit the reality. As per the basic scientific knowledge, the water (H2O) alone is not displaying any electrical properties, i.e. capacitance, unless mixed with ions that transfer the electrical charges from the electrodes of different polarities. The specifications of the commercial epoxy compound list a percentage of ions that can contribute to an electrolytic solution in case of water presence. An electrochemical measure method is therefore more in agreement with the above consideration. The so called Cyclic Stair Case Voltammetry (CSVC) technique applied to the external nodes of the sensor device consist in a regular potential step scanning. The current is sampled just before the subsequent step. The voltage ramp slope is selected in order to reduce the intrinsic capacitive current contribute of the MEMS electrodes and to highlight the electrochemical effect in the resin with a moisture content. The typical hysteresis I-V loop of a non-faradaic effect after the THS cycle will be presented and compared to a flat diagram after the bake. The obtained diagrams show a clear double layer capacitance effect because of the ions mobility in the filler voids regions, water enhanced after the THS step. The equivalent electric circuit describing this chemical phenomenon is the well-known Randles cell. Moreover some experiments on the resin bulk will be reported. The behavior of the resin bulk is different from the local effect measured by the electrochemical method applied to the sensor pads and it will not show the double layer capacitance attributed to the epoxy compound irregular distribution. In conclusion the electrochemical measurements method allows to characterize the electrical behavior of an epoxy molded device affected by moisture absorption and presenting some resin morphological irregularities.|