Influence of mercury speciation and fractionation on bioaccessibility in soils.

Ingestion of contaminated soils by children during hand-to-mouth activities can be a significant exposure pathway to toxic chemicals. Bioaccessibility, which corresponds to the fraction of an ingested contaminant dissolved in the gastrointestinal tract and potentially available for absorption, can be determined by in vitro extractions and gives a conservative value of relative oral bioavailability. The goal of this study was to investigate the validity of the CDM in vitro extraction protocol, developed by Camp Dresser and Mc Kee, by assessing the influence of soil properties and Hg fractionation on bioaccessibility. Mercury bioaccessibility was determined in two pure mercury compounds, two reference materials (a soil and a sediment), and three field-collected contaminated soils. Soils and reference materials were characterized and a sequential extraction procedure was applied to the samples. Bioaccessibility of HgCl(2) was 99.8% in the gastric phase and 88.6% in the intestinal phase, whereas bioaccessibility of HgS was lower than 0.01%. In field-collected soils A, B, M, and, in ERM-CC580, mercury bioaccessibility was lower than 3.2% (below detection). In contrast, CRM 025-050 had a high Hg bioaccessibility (44.3% for gastric phase and 34.7% for intestinal phase). Gastric and intestinal bioaccessibility values were positively correlated with sulfate content in soils (r = 0.99, p < 0.001, for both gastric and intestinal bioaccessibility). In field-collected soils and ERM-CC580, the residual fraction represented near 100% of the mercury recovered, with less than 2% of mercury being in the water-soluble (F1) and CaCl(2)-exchangeable (F2) fractions. In contrast, 46% of mercury in the reference material CRM 025-050 was extracted in the CaCl(2)-exchangeable fraction. Results of the sequential extractions were in agreement with bioaccessibility values, with the sum of the water-soluble and CaCl(2)-exchangeable fractions (F1 + F2) highly correlated with intestinal bioaccessibility values (r = 0.99, p < 0.001). Hence, the sequential extraction procedure used in this study could be a simple means to help validate mercury bioaccessibility.