An assessment of a handheld X-ray fluorescence instrument for use in exploration and development with an emphasis on REEs and related specialty metals
G. J. Simandl,R. S. Stone,S. Paradis,R. Fajber,H. M. Reid,K. Grattan
DOI: https://doi.org/10.1007/s00126-013-0493-0
IF: 5.206
2013-11-17
Mineralium Deposita
Abstract:Handheld (portable) X-ray fluorescence (pXRF) instruments are designed for use in the exploration for base metals, precious metals, and specialty metals (e.g. rare earth elements (REE), Ta, and Nb) and allow rapid decision-making directly in the field. This paper evaluates the technical merits and limitations of pXRF technology in the exploration for specialty metals using data generated from the analysis of three geochemical standards and a silica blank: Standard Reference Material NIST 2780 from the National Institute of Standards and Technology, Gaithersburg; the Certified Reference Material “TRLK” Rare Earth Ore “CGL 124” from the Mongolia Central Geological Laboratory; the Reference Niobium Ore OKA-1 (CANMET); and a silica blank described as Si (IV) oxide (99.8 % on metal basis) from Alpha Aesar (Ward Hill, MA, USA). The data was acquired over a period of nearly 2 years as a by-product of several distinct specialty metal-related projects using the same pXRF instrument and the same settings. Instrumental analytical accuracy was determined using the percent difference (%diff) between the average concentrations of the pXRF instrument readings and the reported certified values of the standard. Percent relative standard deviation (%RSD) was used as a measure of precision. Smaller %diff and %RSD indicate more accurate and precise data, and the accuracy and precision of the pXRF depended strongly on the elemental concentrations in the standards used. Box and whisker diagrams were used to illustrate characteristics of pXRF data sets (mean, lower and upper quartiles, and range) corresponding to individual standards. The bias of the pXRF determinations (under/overestimation) relative to certified values of individual standards are also depicted on these diagrams. This study indicates that the pXRF was capable of producing readings for Si, K, Al, Fe, Ca, Ti, Pb, Zn, Sr, Ag, Cd, Th, Sb, P, S, Mo, Mn, Mg, As, Nb, Rb, La, Ce, Pr, Nd, and Y within 10 %RSD of the reported certified value in at least one of the three standards measured, depending on the concentrations within that standard. The instrument was able to provide relatively accurate and precise data for Nd, Pr, Ce, La, and Y (%diff, ≤17; %RSD, ≤3.7) when the concentrations were greater than 1,000 ppm (e.g., rare earth ore CGL 124). It generated lower quality data (in terms of %diff and %RSD) when concentrations of these elements were in the hundreds parts per million range or lower (NIST 2780). As for Nb, the best %RSD (2.2) was obtained on the Nb Ore OKA-1 with an accompanying %diff of 34. This indicates that the instrument is suitable for use in the exploration and development of carbonatite-related REE deposits such as: Mountain Pass (California), Bayan Obo (Inner Mongolia), Wicheeda Lake (British Columbia), St Honoré, Oka, and Eldor (Quebec). By extension, the instrument is also suitable for the exploration of apatite–monazite veins (Steenkampskraal) and peralkaline intrusion-related deposits such as Kipawa, Nechalacho (Northwest Territories), or Strange Lake (Quebec; Labrador, Newfoundland). A pXRF without a radioactive source cannot be used to analyze for most of the heavy lanthanides; however, Y can be used as a pathfinder for these elements. Limited but effective use of the pXRF in exploration for ion adsorption clay deposits with grades as low as 500 ppm total REE appears possible. The pXRF could also be used in the exploration for carbonatite-related Nb-bearing deposits such as St Honoré (Quebec) and Aley (British Columbia), or in the exploration for lower-grade Nb–Ta deposits such as Upper Fir (British Columbia), justifying the need for future case studies.
geochemistry & geophysics,mineralogy