PRELIMINARY TRACE-ELEMENT DISTRIBUTION USING PXRF GEOCHEMICAL DATA ON AMPHIBOLITE AND PEGMATITE ROCKS FROM SOUTHERN MINAS GERAIS, BRAZIL

Currently, the need for more sustainable, carbon-free alternative energy is rapidly increasing. Within this trend, the development of more efficient electric cars is extremely important, which involves the use of several critical chemical elements, in particular lithium (Li), is widely used in the production of rechargeable Li-ion batteries. A main resource of Li are pegmatites of the lithium-cesium-tantalum (LCT) family. The objective of this work is to validate the use of portable X-ray fluorescence (pXRF) to rapidly quantify Li-related elements in rocks. As Li is a light element, pXRF is not able to detect it, but pXRF can detect and quantify Li-deposit pathfinder elements, such as Cs, Rb, and Sn, etc. This study was conducted on LCT pegmatites from São João del Rei, near Nazareno, Minas Gerais, Brazil. Samples of pegmatite and the surrounding host rock, amphibolite, were analyzed with pXRF. Prior to analysis, a 5x5 cm grid of points was created on a smooth surface for each rock. Measurements collected for this comparative characterization included Al, Fe, Si, Sn, and Rb. The pXRF results were spatialized using the SagaGis software (7.9.0). Heat maps were assembled using ArcMap (10.1). Moreover, the means of the concentration obtained underwent a Tukey Test of 95% of confidence using RStudio (1.4.1106). As expected, Fe, Mn, and Si contents significantly differed between amphibolite and pegmatite rocks. Mn and Rb concentrations at some of the points within the rocks were below detection levels.

Rock	Al	Fe	Rb	Si	Sn
Amphibolite	5.73a	8.42a	0.0007a	18.4b	0.001a
Pegmatite	6.62a	0.04b	0.02a	30.3a	0.001a
* Means in wt.% followed by the same letter in the same column do not differ statistically among themselves by Tukey test (p < 0.05).

This study suggests that pXRF can be used to quickly map trace elements on rock samples or outcrop surfaces and to conduct exploratory screening for LCT pathfinder elements. The next step would be to use pXRF and machine learning algorithms to predict the amount of Li in rocks and soils from various climates, based on Li-related elements.