Improvement of Spark Atomic Emission Spectrometry for Determination of Oxysulfides in Structural Steel

期刊论文
发表日期：

2024年3月1日

查看全文

E. A. Probenkova,
E. V. Yakubenko,
T. N. Ermolaeva,
Yu. N. Orekhova

摘要

Abstract

We have studied conditions of oxysulfide determination in structural steel by spark atomic emission spectrometry (SAES) using an ARL iSpark Model 8860 spectrometer fitted with the Spark-DAT program option for spark diagram processing. In addition, to improve characteristics of calibration curves for determination of Al, Ca, Mn, and sulfur forming nonmetallic inclusions (NMIs), we used structural steel composition standards. The results demonstrate better sensitivity of Al and Ca determination and an increase in correlation coefficients of calibration curves for Mn and sulfur determination. Since no oxysulfide reference standards are available, the percentage of Al₂O₃MnS, Al₂O₃MnSMgO, Al₂O₃MnSCaS, Al₂O₃CaOCaS, Al₂O₃CaOMgOCaS in steel was pre-evaluated in accordance with the ASTM E1245 Standard (method 3) by scanning electron microscopy (SEM) with the use of an energy dispersive electron probe spectrometry (EDS) system. The results indicate that, in SAES determination of oxysulfides in structural steel, it is reasonable to use corrected algorithms for computer processing of the integrated spectrum, in combination with a pseudoformula. A proper algorithm and a pseudoformula for determination of each component were chosen with the use of Student’s t-test statistic by comparing the percentages of NMIs determined by SEM/EDS and SAES. We chose the spark range and analytical signal integration delay time for Al, Ca, Mg, Mn, and sulfur and confirmed the correctness of results of SAES determination of oxysulfides in the range from 500 to 1900 sparks (t_exp = 0.01) at a signal integration delay time of 110 μs (t_exp = 0.23). The technique for SAES determination of the total percentage of oxysulfides was tested in analysis of industrial structural steel samples. The results confirmed that there was no systematic error and that correct determination data were obtained (t_exp < 4.30). The proposed technique for NMI determination allows the analysis time to be reduced from 18 h (SEM/EDS) to 10 min.