Correlative imaging of large-scale geological samples using LIBS and SEM-EDX automated mineralogy

Correlative imaging of large-scale geological samples using LIBS and SEM-EDX automated mineralogy

Článek WoS

Elemental mapping of light (Li, Be, B, F) and trace elements in large-area geological samples represents an analytical challenge. Increased interest in Li and rare-element deposits results in the need for reliable mapping of light and trace elements. One of the possible solutions is laser-induced breakdown spectroscopy (LIBS) combined with sophisticated data processing. We studied four large-area (ca. 8⨯8 cm) samples from two rare-element Be-rich granitic pegmatites with moderate contents of Li + Be (Maršíkov, Czech Republic) and high contents of Li + Be+B (Rau property, Yukon, Canada). Cross-sections from the host rock to the geochemically evolved center of the pegmatite were used for direct multi-element mapping, assisted semi-automated machine-learning-based mineral identification, and tracing of magmatic, metasomatic and hydrothermal processes. From the obtained LIBS data, elemental images of individual elements of interest were produced (incl. Li and Be), and unsupervised clustering to the spectra was applied. This led to cluster images manifesting the distribution of individual matrices (i.e., minerals). The outputs of clustering were visually and statistically compared to a phase map produced by a state-of-the-art SEM-EDS-based Automated mineralogy solution (TIMA) and the results show good agreement. The study demonstrates a very good applicability of LIBS for the detection and mapping of major (Si, Al, Fe, Mg, Mn, Ca, Na), light elements (Li, Be, B, F), and trace elements (Ge, Ga, Cu, Ti, Sr, Ba) in pegmatites, discerning their specific geochemical signatures corresponding to magmatic, metasomatic, and hydrothermal stage of their evolution. The EDS-based automated mineralogy offers a higher spatial resolution in general. Still, it cannot map light elements (H, Li, Be, B) and has between one to three orders of magnitude worse detection limit for many other elements (e.g. Ga…).

Elemental mapping of light (Li, Be, B, F) and trace elements in large-area geological samples represents an analytical challenge. Increased interest in Li and rare-element deposits results in the need for reliable mapping of light and trace elements. One of the possible solutions is laser-induced breakdown spectroscopy (LIBS) combined with sophisticated data processing. We studied four large-area (ca. 8⨯8 cm) samples from two rare-element Be-rich granitic pegmatites with moderate contents of Li + Be (Maršíkov, Czech Republic) and high contents of Li + Be+B (Rau property, Yukon, Canada). Cross-sections from the host rock to the geochemically evolved center of the pegmatite were used for direct multi-element mapping, assisted semi-automated machine-learning-based mineral identification, and tracing of magmatic, metasomatic and hydrothermal processes. From the obtained LIBS data, elemental images of individual elements of interest were produced (incl. Li and Be), and unsupervised clustering to the spectra was applied. This led to cluster images manifesting the distribution of individual matrices (i.e., minerals). The outputs of clustering were visually and statistically compared to a phase map produced by a state-of-the-art SEM-EDS-based Automated mineralogy solution (TIMA) and the results show good agreement. The study demonstrates a very good applicability of LIBS for the detection and mapping of major (Si, Al, Fe, Mg, Mn, Ca, Na), light elements (Li, Be, B, F), and trace elements (Ge, Ga, Cu, Ti, Sr, Ba) in pegmatites, discerning their specific geochemical signatures corresponding to magmatic, metasomatic, and hydrothermal stage of their evolution. The EDS-based automated mineralogy offers a higher spatial resolution in general. Still, it cannot map light elements (H, Li, Be, B) and has between one to three orders of magnitude worse detection limit for many other elements (e.g. Ga…).

Multi-elemental imaging, Clustering, Geology, Automated mineralogy, lithium, Pegmatites

Multi-elemental imaging, Clustering, Geology, Automated mineralogy, lithium, Pegmatites

BUDAY, J.; VÝRAVSKÝ, J.; CEMPÍREK, J.; PROCHAZKA, D.; POŘÍZKA, P.; KAISER, J.

2026

01.01.2026

SPECTROCHIMICA ACTA PART B-ATOMIC SPECTROSCOPY

235

Spojené království Velké Británie a Severního Irska

12

https://www.sciencedirect.com/science/article/pii/S0584854725002824?lid=jua9g5tkojjo&utm_source=braze&utm_medium=email&utm_campaign=STMJ_220042_AUTH_SERV_PPUB&utm_content=07bab9e4-c31e-408b-afa5-1c9ca0f269ca&utm_term=07bab9e4-c31e-408b-afa5-1c9ca0f269ca&DGCID=STMJ_220042_AUTH_SERV_PPUB