Premium
Round Robin on Spectrometer Transmission Calibration for AES in the Common Data Processing System
Author(s) -
Yoshitake M.,
Yoshihara K.
Publication year - 1997
Publication title -
surface and interface analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 90
eISSN - 1096-9918
pISSN - 0142-2421
DOI - 10.1002/(sici)1096-9918(199703)25:3<209::aid-sia228>3.0.co;2-t
Subject(s) - spectrometer , spectrum analyzer , sensitivity (control systems) , optics , computational physics , energy (signal processing) , range (aeronautics) , function (biology) , physics , mathematics , statistics , materials science , electronic engineering , engineering , evolutionary biology , composite material , biology
The results of a round robin on the energy dependence of detection sensitivity are presented in the framework of the shared software, the Common Data Processing System (COMPRO). This paper aims to give a practical way of checking sensitivity for users who use AES as a tool for their material research and are not necessarily interested in the instrument characteristics but are interested in getting more reliable sensitivity factors. After checking the measurement conditions, a wide range spectrum of Cu (0–2000 eV) was obtained by each participant. Each spectrum has been divided by the ‘reference’ spectrum in the COMPRO. The results of division (the ratio function or the spectrometer function) have been analyzed and fitted by the parameters of the electron multiplier. In the case of pulse counting mode, fitting was very good. In some cases, a step has been found in the ratio function around the peak energy, which made the fitting impossible. In order to give a rough estimation of the energy dependence of sensitivity even in such cases, the way to subtract the origin of a step, which may relate to the internal scattered electron in a cylindrical mirror analyzer, has been studied. It was found that when the intensity (which is proportional to the energy resolution Δ E , which is proportional to E ) was subtracted from a raw spectrum intensity, the ratio function calculated from the subtracted spectrum and the ‘reference’ spectrum had a shape which can be fitted by the parameters of the electron multiplier. It also becomes apparent that the misalignment of the sample position or the beam position leads to the wrong spectrometer function and that it is very important to take the spectrum under the optimum condition in order to obtain a reasonable spectrometer function. The routine to divide a spectrum, to load the ‘reference’ spectrum, to fit the spectrometer function with parameters and to calibrate a sample spectrum with the fitting parameters are all supplied in the COMPRO. © 1997 by John Wiley & Sons, Ltd.