X-ray fluorescence spectrometry and electronic microprobe analysis - Complementary techniques for precision measurement of trace chlorine in volcanicglass formed in mid-oceanic ridges

Citation
J. Etoubleau et al., X-ray fluorescence spectrometry and electronic microprobe analysis - Complementary techniques for precision measurement of trace chlorine in volcanicglass formed in mid-oceanic ridges, J PHYS IV, 8(P5), 1998, pp. 445-452
Citations number
18
Categorie Soggetti
Physics
Journal title
JOURNAL DE PHYSIQUE IV
ISSN journal
11554339 → ACNP
Volume
8
Issue
P5
Year of publication
1998
Pages
445 - 452
Database
ISI
SICI code
1155-4339(199810)8:P5<445:XFSAEM>2.0.ZU;2-A
Abstract
The relative abundance of chlorine (Cl) in silicate rocks, point out this e lement as a potential geochemical indicator of first importance. However th e contents of this halogen in geological sample are poorly known, mainly du e to the difficulty to measure efficaciously this element present in most c ases at trace level. In this work, X-ray fluorescence spectrometry and elec tron microprobe are closeness associated to determined precisely chlorine i n volcanic glasses sampled on mid-oceanic ridges. X-ray fluorescence is use d first to certify chlorine amounts in a set of volcanic glasses to calibra te the electron microprobe. X-ray fluorescence determination is made accord ing to an original procedure. Chlorine, volatile element, is eliminated fro m the finely rock powder sample by pyrolise. A calibration curve is realize d with press pellets of calcinated rock powder and doped with increasing am ounts of a KCI standard solution. Chlorine concentration in sample is measu red on a press pellet prepare with un-calcinated rock powder and read on th is calibration curve. Chlorine concentrations found in X-ray fluorescence a re linearly correlated with microprobe intensities measured on a thin secti on of small separate glass ships from the same rock samples. Because chlori ne volatitility, it is necessary to define judiciously electron microprobe measurement conditions. Optimization of analytical parameters and repetitio n of measurement cycles allow to obtain results with a detection limit lowe red to 26 ppm and a precision (standard deviation) better than 30 ppm in 0- 700 ppm concentration range.