Fused beads for the analysis of Li bearing samples by LIBS and XRF

In recent decades, it has become apparent that due to the increasing complexity of deposits, the complexity of ore samples is also increasing. At the same time, the need for detailed geological and mineralogical information also increased. Therefore, it is important to combine methods in order to obtain more comprehensive conclusions.
We present a method of combining WDXRF (wavelength dispersive X-ray fluorescence) analyses with a wide spectrum of elements and pLIBS (portable laser induced breakdown spectroscopy, see Fig. 1) with a likewise spectrum and the additional possibility to detect light elements such as Li. Nevertheless, the representativeness of LIBS analyses is significantly smaller compared to XRF. To overcome this and effectively combine both methods, we designed and produced fused beads using Na2B4O7 as flux. Unfortunately, such beads are transparent to the laser which makes LIBS analysis almost impossible, therefore, we tested two different approaches: 1) dying the bead by adding CuO to the flux and 2) roughen the beads surface. We used a mixture of Na2B4O7 (+ CuO) + KI as matrix. Potassium iodide (KI) was added as a releasing agent. We used Greisen rocks from the Altenberg-Zinnwald district with known composition – including Li – and LiBO2 in different concentration ratios to adjust the Li content.
To produce crack-free beads, we had to adapt the melting process in the fully automatic melting furnace TheOx (CLAISSE). For the analytical work, we use a portable LIBS instrument manufactured by SciAps (Z-300, laser wavelength 1064 nm) and a PANalytical Axiosmax minerals XRF spectrometer.
The fused beads will now serve as calibration samples for both XRF and LIBS measurements (see Fig. 2). Samples with unknown composition will be analysed by XRF first. It turned out that roughening the beads has no significant influence on the XRF-spectra as revealed in before and after scan measurements. In a second step the LIBS analysis is performed in particular for light element determination such as Li. Furthermore, adding a dying component such as Cu might be suitable as a robust internal standard for both LIBS and XRF.

The obvious disadvantage of the method is that neither Cu, K nor Na can be included in the quantitative analysis. Depending on the analytical problem, they would have to be determined in a preceding step (e.g. via a conventional Li-borate melt and XRF).