Development of a multi-variate calibration approach for quantitative analysis of oxidation resistant Mo–Si–B coatings using laser ablation inductively coupled plasma mass spectrometry

Abstract

Nowadays, for the production of oxidation protection coatings in ultrahigh temperature environments, alloys of Mo–Si–B are employed. The properties of the material, mainly the oxidation resistance, are strongly influenced by the Si to B ratio; thus reliable analytical methods are needed to assure exact determination of the material composition for the respective applications.For analysis of such coatings, laser ablation inductively coupled mass spectrometry (LA-ICP-MS) has been reported as a versatile method with no specific requirements on the nature of the sample. However, matrix effects represent the main limitation of laser-based solid sampling techniques and usually the use of matrix-matched standards for quantitative analysis is required. In this work, LA-ICP-MS analysis of samples with known composition and varying Mo, Si and B content was carried out. Between known analyte concentrations and derived LA-ICP-MS signal intensities no linear correlation could be found. In order to allow quantitative analysis independent of matrix effects, a multiple linear regression model was developed. Besides the three target analytes also the signals of possible argides (40Ar36Ar and 98Mo40Ar) as well as detected impurities of the Mo–Si–B coatings (108Pd) were considered. Applicability of the model to unknown samples was confirmed using external validation. Relative deviations from the values determined using conventional liquid analysis after sample digestion between 5 and 10% for the main components Mo and Si were observed.