A framework is presented for considering the relative strengths of Auger electron spectroscopy (AES)/scanning Auger microscopy (SAM) and scanning transmission electron microscopy–electron energy loss spectroscopy (STEM-EELS) when selecting a defect analysis technique. The geometry of the analysis volumes for each technique is visualized. The analysis volume for AES/SAM is shaped like a button while the STEM-EELS analysis volume is more like a thread extending throughout the thickness of the prepared sample. The usefulness of this framework is illustrated with the example of small defect particles. In this example the size and shape of the AES/SAM analysis volume is a better fit to the defect, thus it provides better chemical analysis while STEM provides better images of the defects.

Shallow junction formation in silicon chips is a hot topic in the semiconductor industry. Reduction of power consumption of integrated circuits and an increase of the device performance would drastically reduce the sizes of circuits and, therefore, would necessitate a similar reduction for the depth of the p-n junctions. This paper performed the flash lamp annealing process application, applied to the CMOS as alternative method to attain the goal of shallow junction in the case. A marginal thermal budget mismatch related failure mode was revealed and explained.

In this paper, a case of package level reliability test failure was studied. A model of “Slice Defect”, which was identified as the root cause by failure analysis, is introduced. Experiment results are presented to approve that such model is in fact correct and the corrective actions are effective.

Multiple parts failed during a 96 hour HAST (highly accelerated stress test) run. Electrical failure occurred on several pins stressed at 48V during the run. Visual inspection identified possible corrosion damage occurring on a top layer aluminum metal line linked to the failed pins. Additionally, significant lengths of this line and metallization at six other sites appeared white and reflective when viewed through an optical microscope. The device technology utilized a TiN ARC. Aluminum metal with a TiN ARC has a dull, amber color when viewed through an optical light microscope, as opposed to bare aluminum, which appears white and shiny. The initial assumption was that the passivation had lifted off during mold compound removal, along with the top TiN ARC layer at these seven locations. SEM inspection found that final passivation film was still intact over these shiny Al lines, but it was cracked extensively. Neighboring Al lines did not show cracked passivation. A hypothesis was generated that suggested that the TiN ARC was not removed, but rather was altered in some way so as to change its optical appearance. The change in the TiN was believed to be due to a combination of factors that resulted from electrical overstressing of the lines during HAST. A series of experiments utilizing FIB cross-sections, Auger mapping, Auger depth profiling, TEM inspection and EDS were used to show that the TiN ARC layer was still present on the affected lines but had been oxidized. The conclusions drawn from this investigation can be used to rapidly determine the root cause of failure through signature analysis. Shiny Al metal lines are easy to see with optical microscopes and are therefore a useful failure analysis tool to identify electrically and mechanically overstressed lines and circuits.