Problems with elements not included in the software of handheld XRF analysers
Research output: Contribution to conference › Abstract › peer-review
Authors
Organisational units
Abstract
Problems with elements not included in the software of handheld XRF analysers
or
Commercial XRF software, what to do when crucial elements are not included
A Rachetti, K Stocker, W Wegscheider, J Raith
Small portable XRF analysers are widely used for prospecting in the mining industry and plenty commercial applications are explicitely advertised by the manufacturers. Unfortunately the analyst is depending on the software to include all potentially interfering elements.
In this work we explore the effect of a neglected element, yttrium, on the quantification of the element of interest, niobium, in the analysis of stream sediments and granite eluvium of different concentrates. The samples stem from the Bohemian Massif (Austria) and are investigated in order to verify known geochemical anomalies and their potential use as prospectivity indicators for coltan deposits within this area.
The first strategy to account for yttrium not in calibration was to fit the spectra with the freely available software PyMca [http://pymca.sourceforge.net/] and to use the resulting net intensities of the element peaks for a multivariate calibration. Correct concentrations for yttrium and other elements were supplied by independent ICP MS analyses. Advantages and limitations of this approach will be discussed. This approach was favoured as samples representing intermediary steps of preparation were not available for a second measurement anyway.
The second strategy was to send the instrument back to the manufacturer and ask for a recalibration including the missing element. This was rated second best as the remeasurement of 700 + samples was either not possible and/or a tedious affair. Good thing that the manufacturer showed consideration for time lost and did the software update for free.
Data from before and after recalibration and the data from the multivariate approach will be compared.
( Special thanks go to Felix Meixner, who helped his mother with assembling endless batch runs of PyMca and wrote a small program to automatically convert the output to Excel.)
or
Commercial XRF software, what to do when crucial elements are not included
A Rachetti, K Stocker, W Wegscheider, J Raith
Small portable XRF analysers are widely used for prospecting in the mining industry and plenty commercial applications are explicitely advertised by the manufacturers. Unfortunately the analyst is depending on the software to include all potentially interfering elements.
In this work we explore the effect of a neglected element, yttrium, on the quantification of the element of interest, niobium, in the analysis of stream sediments and granite eluvium of different concentrates. The samples stem from the Bohemian Massif (Austria) and are investigated in order to verify known geochemical anomalies and their potential use as prospectivity indicators for coltan deposits within this area.
The first strategy to account for yttrium not in calibration was to fit the spectra with the freely available software PyMca [http://pymca.sourceforge.net/] and to use the resulting net intensities of the element peaks for a multivariate calibration. Correct concentrations for yttrium and other elements were supplied by independent ICP MS analyses. Advantages and limitations of this approach will be discussed. This approach was favoured as samples representing intermediary steps of preparation were not available for a second measurement anyway.
The second strategy was to send the instrument back to the manufacturer and ask for a recalibration including the missing element. This was rated second best as the remeasurement of 700 + samples was either not possible and/or a tedious affair. Good thing that the manufacturer showed consideration for time lost and did the software update for free.
Data from before and after recalibration and the data from the multivariate approach will be compared.
( Special thanks go to Felix Meixner, who helped his mother with assembling endless batch runs of PyMca and wrote a small program to automatically convert the output to Excel.)
Details
Original language | English |
---|---|
Publication status | Published - 2015 |
Event | Geoanalysis2015 - Leoben, Austria Duration: 9 Aug 2015 → 13 Aug 2015 |
Conference
Conference | Geoanalysis2015 |
---|---|
Country/Territory | Austria |
City | Leoben |
Period | 9/08/15 → 13/08/15 |