Monitoring Mercury in size fractionated road dust samples by flow injection-cold vapor-atomic absorption spectrometry. Where is Hg coming from?

LONDONIO AGUSTIN; FABIÁN FUJIWARA; RAUL REBAGLIATI; DARÍO GÓMEZ; PATRICIA SMICHOWSKI

BENTO GONÇALVES

Encuentro; 3º Encontro Brasileiro sobre Especiação Química; 2012

UFSM

An investigation was carried out to explore the effect of citric, hydrochloric, oxalic and tartaric acids on Hg vapor generation of in a flow injection system in conjunction with atomic absorption spectrometry detection (CV-AAS) for its determination in road dust samples [1]. Samples were collected in Buenos Aires (Argentina) during two months at 15 sites grouped in five zones with different urban characteristics and traffic profile. They were fractionated in four fractions (with the following particle diameter (d): A: d < 37 µm, B: 37 < d < 50 µm, C: 50 < d < 75 µm, and D: 75 < d < 100 µm) before Hg determination. The operating conditions (chemical and physical parameters) and the type and concentration of the acids studied (citric, hydrochloric, oxalic and tartaric) were assessed for the efficient generation of Hg. The interfering effect of metals and metalloids, present as minor and major concentrations in road dust, on Hg signal were evaluated using the four acids. Hydrochloric acid resulted to be the most convenient reaction medium in terms of sensitivity but a suitable control of interferences was not reached. For a complex matrix such as road dust, the best control of interferences was achieved when oxalic acid was tested, even when lower Hg signal were obtained in comparison with HCl. The detection limit (3σ criterion) turned out to be 0.3 ng g-1. For checking accuracy, two certified reference materials were analyzed and results were in good agreement with certified values. Mercury concentrations in road dust ranged between 0.1 µg g-1 and 4.6 µg g-1 and the mean concentration value resulted to be 1.09 µg g-1. In general, the finer fractions (A, B) exhibited higher Hg concentrations than the coarser ones (C, D), with average concentrations (µg g-1) in the order: A (1.5) ~ B (1.6) > C (0.6) = D (0.6). Mercury data obtained were analyzed statistically together with the other 14 elements measured in the same samples to find a common pattern. Mercury correlated with S and Zn (present in liquid fuels). No significant correlations were detected between Hg and other traffic related elements such as Cd, Cu, Mo, Pb and Sb that are mostly associated with mechanical wear. These characteristics allow us to link the presence of Hg in Buenos Aires road dust with exhaust of liquid fuel combustion from vehicles