Premium
The Bulk Analysis of Silicate Rocks by Portable X‐Ray Fluorescence: Effect of Sample Mineralogy in Relation to the Size of the Excited Volume
Author(s) -
Potts Philip J.,
WilliamsThorpe Olwen,
Webb Peter C.
Publication year - 1997
Publication title -
geostandards newsletter
Language(s) - English
Resource type - Journals
eISSN - 1751-908X
pISSN - 0150-5505
DOI - 10.1111/j.1751-908x.1997.tb00529.x
Subject(s) - analyser , silicate , analytical chemistry (journal) , mineralogy , x ray fluorescence , standard deviation , grain size , penetration depth , petrography , volume (thermodynamics) , relative standard deviation , chlorite , excited state , materials science , fluorescence , chemistry , optics , physics , detection limit , statistics , mathematics , quartz , atomic physics , metallurgy , organic chemistry , chromatography , quantum mechanics
This paper examines the limitations arising from sample mineralogy when a portable X‐ray fluorescence instrument is applied to the direct in situ analysis of silicate rocks. Estimates were made of the size of the excited volume from which the X‐ray fluorescence signal originates by calculating the critical penetration depth for selected X‐ray lines. Measurements were made of the variations in detected intensities over the area of the P‐XRF analyser window and showed that, using radioisotope sources incorporated in the instrument used in this study (Spectrace TN9000), excitation intensities were six to ten times greater at the centre compared to the edge of the analyser window. These data indicated that the region of the sample at the centre of the window will make an enhanced contribution to detected spectra. Replicate measurements on slabs of rock selected to represent fine‐to coarse‐grain size textures indicated the magnitude of the sampling precision that can be achieved in the direct analysis of silicate rocks. Typical values were better than 5% relative standard deviation of the mean from an average of five determinations on fine‐ to medium‐grained rocks and better than 10 % relative standard deviation in a single determination on these samples.