Abstract

The use of X-ray fluorescence spectrometry (XRFS) on fusion discs to determine Fe in iron ores provides high quality data and is widely used by iron ore producers, but is costly. This method of sample preparation eliminates mineral and particle size effects which can lead to biased results when using pressed powder pellets. For many applications, such as exploration and mining control, the method of the XRFS analysis of iron ores by pressed powder pellet probably provides data of adequate quality and this method is now used for most analyses by Kumba Iron Ore, an iron ore producer in South Africa. Automation of sample preparation procedures should improve the quality of data for Fe, K2O and P determined by XRFS using pressed powder pellets. This study was carried out by Kumba to compare the quality of data obtained using the three different sample preparation methods and using precision as the measurement of quality. The three methods were (i) automated sample preparation, measurement on fused discs, (ii) automated sample preparation, measurement of pressed powder pellets, and (iii) manual sample preparation, measurement of pressed powder pellets. It was found, as expected, that the precision was best for the measurements on fusion discs where automation was used for sample preparation procedures. The relative standard deviations (RSDs) for these for Fe, K2O and P were 0·074, 1·906 and 1·636% respectively. However, the XRFS data generated from the pressed powder pellets, where automation was used for sample preparation procedures, compare remarkably well against those from the auto-prepared fusion discs. The RSDs for Fe, K2O and P for these were 0·131, 2·912 and 1·989% respectively. As expected, the poorest quality data resulted from the XRFS measurements on pressed powder pellets where manual methods only were used for sample preparation. The RSDs for Fe, K2O and P for these were 0·194, 4·361 and 3·430% respectively. It is concluded that the analysis of iron ores by pressed powder pellet can produce acceptable quality analytical data for many applications. Also the introduction of automation for sample preparation in the iron ore analytical laboratory does improve the quality of data produced by this XRFS analytical method.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call