Abstract <p>A method for determining the carbon content of carbonate–aluminosilicate glasses by electron probe microanalysis is proposed and implemented. To take into account the C<i>K</i><sub>α</sub> signal arising due to the conductive carbon coating and beam-induced carbon contamination, we used a calibration approach with carbonate reference samples of different carbon concentrations. The analytical signal was determined as the area under the first C<i>K</i><sub>α</sub> peak in the recorded X-ray spectrum, which ensured the account of the effects of carbon–oxygen bonding and interactions with other elements on the C<i>K</i><sub>α</sub> peak shape and position. Optimal conditions for recording spectra in the C<i>K</i><sub>α</sub> region were found for accurate carbon determination. Linear calibration curves obtained at 5 kV using carbonate standards were highly reproducible, with relative standard deviations of 1–3%. The method ensures limits of detection of 0.2–0.4 wt % and limits of quantification of 0.6–1.2 wt %.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Determination of Carbon in Carbonate–Silicate Glasses by Electron Probe Microanalysis

  • A. A. Viryus,
  • V. Yu. Chevychelov

摘要

Abstract

A method for determining the carbon content of carbonate–aluminosilicate glasses by electron probe microanalysis is proposed and implemented. To take into account the CKα signal arising due to the conductive carbon coating and beam-induced carbon contamination, we used a calibration approach with carbonate reference samples of different carbon concentrations. The analytical signal was determined as the area under the first CKα peak in the recorded X-ray spectrum, which ensured the account of the effects of carbon–oxygen bonding and interactions with other elements on the CKα peak shape and position. Optimal conditions for recording spectra in the CKα region were found for accurate carbon determination. Linear calibration curves obtained at 5 kV using carbonate standards were highly reproducible, with relative standard deviations of 1–3%. The method ensures limits of detection of 0.2–0.4 wt % and limits of quantification of 0.6–1.2 wt %.