The challenges of Li determination in minerals: a comparison of stoichiometrically determined Li by EPMA with direct measurement by LA-ICP-MS and handheld LIBS.

Lithium ores are potentially complex, including as Li-bearing phases lepidolite, the amblygonite-montebrusite group, lithiophosphate(tr) and petalite, and because its low atomic mass precludes direct electron-beam measurement analysing the Li content of whole rock is not straightforward with more than 10% difference sometimes noted between methods; an example compares spodumene pegmatite sampes from Kaustinen, Finland, analysed by 4-acid digestion versus Na2O2 flux then acid, both with ICP-AES finish. Whole-rock data are given from Kaustinen, the Goncalo lepidolite/Li phosphates in Portugal and the Cinavec Li-micas in Bohemia, advice on whole-rock procedure is summarised and cross-sections are presented of the Cinovec deposit's petrology, Sn and Li deposit contents and grade distribution block modelling; the Czech deposit is the largest low-cost hard-rock Li resource in Europe with 7 000 000 tonnes indicated and inferred lithium carbonate equivalent resources at 0.1% Li cut-off, contained in an indicated 347 700 000 t at 0.45% Li2O, 0.04% Sn and an inferred 308 800 000 t at 0.39% Li2O, 0.04% Sn; the exploration target is 350 000 000-450 000 000 t at 0.39-0.47% Li2O. The mineralogical characterisation of the various Li minerals is illustrated and compared for EPMA, from which the values have to be recalculated using stoichiometry, and direct measurement by LA-ICP-MS (laser ablation-inductively coupled plasma-mass spectrometry), showing that the majority of minerals have a good analytical agreement between the methods but that lithiophosphate and bityite results both differ significantly while hectorite and jadarite give variable results. LIBS (laser-induced breakdown spectroscopy) gives data that has a reasonable correlation with the EPMA and ICP-MS data, considering the spatial sampling differences.
Lithium ores are potentially complex, including as Li-bearing phases lepidolite, the amblygonite-montebrusite group, lithiophosphate(tr) and petalite, and because its low atomic mass precludes direct electron-beam measurement analysing the Li content of whole rock is not straightforward with more than 10% difference sometimes noted between methods; an example compares spodumene pegmatite sampes from Kaustinen, Finland, analysed by 4-acid digestion versus Na2O2 flux then acid, both with ICP-AES finish. Whole-rock data are given from Kaustinen, the Goncalo lepidolite/Li phosphates in Portugal and the Cinavec Li-micas in Bohemia, advice on whole-rock procedure is summarised and cross-sections are presented of the Cinovec deposit's petrology, Sn and Li deposit contents and grade distribution block modelling; the Czech deposit is the largest low-cost hard-rock Li resource in Europe with 7 000 000 tonnes indicated and inferred lithium carbonate equivalent resources at 0.1% Li cut-off, contained in an indicated 347 700 000 t at 0.45% Li2O, 0.04% Sn and an inferred 308 800 000 t at 0.39% Li2O, 0.04% Sn; the exploration target is 350 000 000-450 000 000 t at 0.39-0.47% Li2O. The mineralogical characterisation of the various Li minerals is illustrated and compared for EPMA, from which the values have to be recalculated using stoichiometry, and direct measurement by LA-ICP-MS (laser ablation-inductively coupled plasma-mass spectrometry), showing that the majority of minerals have a good analytical agreement between the methods but that lithiophosphate and bityite results both differ significantly while hectorite and jadarite give variable results. LIBS (laser-induced breakdown spectroscopy) gives data that has a reasonable correlation with the EPMA and ICP-MS data, considering the spatial sampling differences.