T132 Critical Metals (REE, In, Te, Nb, Ta, Ga, Li, etc.) for the New Energy Future

Monday, 1 November 2010
1:30 PM - 5:30 PM
Room 601

PROVENANCE DETERMINATION OF CONFLICT MINERALS BY LASER-INDUCED BREAKDOWN SPECTROSCOPY: THE EXAMPLE OF COLUMBITE-TANTALITE

Paper number: T132-10
Program: Technical Sessions
Time: 3:45 PM – 4:00 PM

Authors

Abstract

The primary natural occurrence of Nb and Ta is in the complex oxide minerals columbite and tantalite, which form a solid-solution series with the general composition (Fe,Mn)(Nb,Ta)2O6. Central Africa is a major source for columbite-tantalite, which is included in the suite of ‘conflict minerals’ whose illicit export is considered responsible for financing the ongoing civil conflicts in this region. With the passing into US law of the financial reform bill of 2010 (S.3217 - Restoring American Financial Stability Act of 2010), US companies will soon be required to report annually to the Securities and Exchange Commission about whether their products contain gold, tin, tungsten, and tantalum from the Democratic Republic of Congo or adjacent countries. Thus, being able to recognize the ‘conflict minerals’ that are the source for these elements has taken on increased urgency.

Laser-induced breakdown spectroscopy (LIBS) may be a convenient and low-cost means for rapidly determining a mineral’s geographic origin because the LIBS plasma emission spectrum provides a ‘chemical fingerprint’ of any material in real-time. Toward this end, five sets of columbite-tantalite samples from spodumene-bearing granite pegmatites in Maine, California, and the Northwest Territories of Canada have been analyzed. Individual spectral emission lines and line ratios for Fe, Mn, Nb, Ta, W, Ti, Sc, Sn, and Zr were measured using an RT100-HP LIBS instrument (A3 Technologies LLC/Applied Spectra, Inc.) employing a 90mJ, 1064 Nd:YAG laser and CCD detector. Both broadband (250-490 nm) and narrowband (240-280 nm and 315-355 nm) spectra were used to determine if the provenance of the samples could be discriminated using advanced multivariate statistical techniques. Partial Least Squares Discriminant Analysis (PLS-DA) gave 100% sample-level classification using a highest confidence approach. These promising preliminary results suggest that additional work with a larger, more geographically diverse data set that includes samples from the major production areas in Australia, Brazil, China, Ethiopia, and Mozambique is warranted.