What is the eV line? is it K-alpha 15.744 kV or 2.042 L-alpha kV you may be getting a Si (1.739 kV) off set interference from aluminum base or other alloy element.

Dear Edward, thanks a lot for your reaction. Both intensities are present in the spectra, and the latest results showed that a similar Zr peak was detected when Mg2Si eutectic was analysed. It is clear that the Mg2Si lamellas nearly fully oxidised during electrolytic thinning, and even you can't get an appropriate diffraction pattern from Mg2Si because of the amorphisation. But under these circumstances, the intensity of Zr is still present in the spectrum.

Well if both peaks for Zr are there then Zr is there - maybe grinding media (ZrO) contamination.

Dear Mykhalenkov, I will be performing the TEM analysis of as-cast Al alloys soon and will let you know if I encounter this problem.

Dear Safir, thanks a lot for your answer. I understood that the strong emission line of Zr in EDS spectra originated not from the specimen but from the environment. Since Zr Ka and Kb have no competitors to be overlapped and Zr content inside the Al matrix measured was as high as 0.8 wt.%, it is clear that the origin of Zr is not inside the alloy. But from where? - that was the reason for my question. I will be glad to discuss this topic further. Some years ago, I also was in Soth Korea as visiting scientist. It was in Changwon station, and the issue was to design Al-Mg-Si-Mn casting alloy with excellent mechanical properties. Right now my cooperator is retired and I have no contacts in Korea.