Impact of ancient metal smelting on arsenic pollution in the Pecora River Valley, Southern Tuscany, Italy.

A significant As anomaly has been reported in the literature for stream sediments and unlithified Quaternary deposits of the Pecora River valley in Southern Tuscany, extending from the “Colline Metallifere” pyrite-base metals district to the Tyrrhenian Sea. The As anomaly spreads over several square kilometers around a core that exceeds 500 ppm. Several source contributions (from natural to anthropogenic) have been invoked to explain the observed As distribution in the Pecora Valley, including the metal-working industry which was active in this area, particularly in Etrusco-Roman times and in the Middle Ages. In order to evaluate the contribution of ancient mining and metallurgical activities in the Pecora Valley to elevated As concentrations in the environment, a detailed mineralogical and geochemical survey of metallurgical slags and smelted ore minerals was undertaken from six different sites through the Pecora Valley: Poggio Butelli (Etrusco-Roman iron slags); Sata Creek, Arialla, Marsiliana, Forra and Cascata sites (all Medieval base metals slags). The As content of Etrusco-Roman slags is relatively low (few tens of ppm), whereas Medieval slags show variable, but higher amounts of base metals (±Ag) (ranging from tens to tens of thousands ppm) and As (up to 267 ppm, with average contents of about 40 ppm). Arsenic is mostly partitioned in sulfides disseminated through the glassy groundmass rather than in solid solution with the glassy matrix. Remnants of the ore used for base metal and Ag smelting during the Middle Ages had the highest As contents (up to about 1000 ppm). The overall As concentration in the slags, independent of age and type, is generally lower than that found in stream sediments and Quaternary alluvial deposit outcrops of the upper Pecora Valley. Textural observations indicate that exogenous alteration of slags is minor and mostly confined to superficial exposures of matte droplets or along fractures; metal particles embedded within the glassy groundmass, even those located a few micrometres from the surface, appear to be unaltered. It is suggested, therefore, that As and contaminant metal contributions to the Pecora valley by archaeometallurgical slags was minor, at least for slags from the last 2500 to 700 a BP. Similarly, based on gross mass balance calculations, airborne release of As due to smelting/roasting operations in the past cannot account for the observed extent of the As anomaly in stream sediments and Quaternary deposits. Although residues of unsmelted (or partially smelted) Fe–Cu–Pb–Zn charges or fluxes for base metal metallurgy contain much greater amounts of As than smelting slags, their relatively low abundance would limit the overall input of As into the Pecora river valley. On the other hand, concurrent research seems to indicate a major “geogenic” contribution to the As anomaly in the Pecora Valley.