Determination of Hg0 in solid matrices: a new approach with Lumex-Pyro thermal desorption. The case study in the former mine of Abbadia San Salvatore (Mt. Amiata, Southern Tuscany, Italy)

The variability of chemical-physical, toxicological, and bioavailability characteristics diversifying Hg species (volatile metallic, insoluble sulfide, soluble salts, organic forms) is a factor that conditions the study of contamination of environmental matrices and related environmental and health risks. Consequently, it influences the design of interventions for the reclamation of contaminated sites and the evaluation of the outcomes of such interventions. For these reasons, within the monitoring activities carried out at the former mining site of Abbadia San Salvatore (Mt. Amiata, Southern Tuscany), where a remediation project is underway, an analytical method aimed at discriminating the content of Hg0 from HgS (cinnabar), which was the mineral cultivated and processed in the mining plants, has been developed for solid matrices. The method uses the Lumex RA-915M instrument as a detector through atomic absorption spectrophotometry (AAS) and a pyrolyzer (Lumex Pyro-915+). The combined use of this instrumentation allows the quantification of the various Hg species in previously ground and homogenized solid samples, differentiating them by applying a temperature ramp set taking into account the different volatilization temperatures of the species themselves. The pyrolyzer consists of two separate cells: the first heats the sample, causing the release of Hg vapours, while the second is heated to approximately 800°C to prevent contamination. The desorption temperature in the sample holder is measured with a thermocouple that monitors the effects of the voltage applied to the pyrolyzer. The air is continuously drawn by the instrument pump, filtered with activated carbons, and introduced into the measurement cell which is exposed to UV radiation emitted by a Hg lamp. The temperature ramp consists of five steps, differentiated by increasing voltage values set independently to the three cells,with a sample processing duration of approximately 15’. The pump flow rate can be placed in the range of 1÷3 L/min, depending on the presumed concentration of Hg in the sample. A straight line on three points was built for the calibration, using increasing amounts of a certified reference material (Montana soils 2711a). The method was tested on samples of cinnabar, calcines, sediments, and soils collected at the Abbadia S. Salvatore mine. To observe the effect of different matrices on thermal desorption, cinnabar was mixed with calcite and synthetic silica. The method allows Hg speciation, avoiding complex sample preparation procedures. However, in order to optimize working conditions, a preliminary mineralogical characterization is desirable. Further benefits could be achieved by implementing temperature control in the cells, avoiding the use of an external thermocouple.