Chemical state of arsenic and copper in enargite: evidences from EPR and X-ray absorption spectroscopies, and SQUID magnetometry

The chemical state of elements in enargite, Cu(3)AsS(4), has been traditionally interpreted assigning the monovalent (+1) state to copper, pentavalent (+5) to arsenic, and divalent (-2) to sulphur. However, there are some evidences that make this interpretation not fully satisfying. These uncertainties reflect on the understanding of the oxidation process of enargite, that has important implications for both environmental issues and mineral processing. To clarify the issue, our study involves the combined use of X-ray absorption (XAS) and electron paramagnetic resonance (EPR) spectroscopies, and SQUID magnetometry. XANES spectra at the Cu K-edge for enargite and other copper sulphides show the expected affinities between the spectra of enargite and chalcopyrite (although the pre-edge feature typical of chalcopyrite is absent in enargite), reflecting the structural similarities of these two minerals; the spectra are, on the other hand, fairly similar to tetrahedrite, whose structure is different. The spectral shapes of bornite, chalcocite, and covellite are distinctly different from enargite. The edge position of enargite is consistent with the presence of monovalent copper, in spite of some small differences with respect to other sulphides where Cu is considered monovalent. At the As K-edge, the most striking feature is the clear difference in the edge position of enargite with respect to As(V) compounds such as calcium arsenate, and the similarity with As(III) compounds such as sodium arsenite. On the other hand, the edge position for enargite occurs at higher energies than those of tennantite, where As is regarded as trivalent, and of realgar (As(4)S(4)), thus supporting the view that in enargite arsenic is in a higher valence state. Enargite does not show a recognizable EPR spectrum in the X band region; however, the mineral shows a rather strong bulk magnetisation, equivalent (in the hypothesis of divalent Cu) to 0.05 unpaired electron per formula unit. The most likely explanation of this behaviour is the occurrence of small amounts of clustered paramagnetic impurities (e. g., divalent Fe). In summary, to the extent that nominal valence states can describe complex chemical bonding such as in sulphides, the results of this study rule out the occurrence of a significant fraction of Cu(II) in enargite, and support the "traditional" assignment of a monovalent state to Cu, and pentavalent to As.