Premium
Round‐robin layer‐thickness determination: Toward reliable reference‐free X‐ray spectrometry
Author(s) -
Sakurai Kenji,
Kurokawa Akira
Publication year - 2018
Publication title -
x‐ray spectrometry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.447
H-Index - 45
eISSN - 1097-4539
pISSN - 0049-8246
DOI - 10.1002/xrs.2978
Subject(s) - round robin test , calibration , reproducibility , spectrometer , x ray fluorescence , analytical chemistry (journal) , materials science , mass spectrometry , x ray , software , layer (electronics) , fluorescence , computer science , chemistry , optics , physics , nanotechnology , chromatography , mathematics , statistics , quantum mechanics , programming language
The present paper reports the results of a round‐robin test recently achieved in Japan for X‐ray fluorescence thickness determination of multilayered samples, such as gold/nickel/copper on a chromium coated quartz glass substrate. In the latest practical application of X‐ray fluorescence analysis, reference‐free determination, which does not rely on the use of an experimentally prepared calibration curve, is frequently employed. In addition, most of the analysis is quite often done by nonexperts, just with the aid of computer software. Therefore, it is extremely important to examine the reliability of reference‐free X‐ray fluorescence determination, compared with chemical analysis. Eleven companies participated in the present round‐robin test and provided 15 datasets in total. The participants measured common samples by using a variety of setups, including both energy‐dispersive and wavelength‐dispersive spectrometers, with different X‐ray tubes (Rh, Pd, Mo, W, and Ag) and power settings, with different filtering, etc., freely chosen and set by themselves. It has been found that X‐ray spectrometry can give stable and reliable values, without any apparent dependence on the types of the X‐ray spectrometers, experimental conditions, and software. The average of the reported mass thickness matches well with chemical analysis, and the standard deviation of the reported thickness was 4.3–6.6%. One of the most fruitful conclusions of the present round‐robin test is that the reproducibility in obtaining the layer‐thickness is good enough to help calibration and confirmation of X‐ray spectrometers currently working in any workplaces, by measuring the same samples and checking the obtained thickness.