| Abstract
| - At As-contaminated sites, where the ingestion of soil bychildren is typically the critical human-health exposurepathway, information on the bioavailability of soil-bound Asis often limited. The influence of various soil physicaland chemical properties (iron and manganese oxides, pH,cation exchange capacity, total inorganic and organiccarbon, and particle size) on As(III) adsorption, sequestration,bioaccessibility (as a surrogate for oral bioavailability),and oxidation was investigated in 36 well-characterized soilsby use of a physiologically based extraction test (PBET).These results were compared to an earlier published studywith As(V) on the same set of soils. The properties ofthe soils were able to explain >80% of the variability inthe adsorption and sequestration (as measured by thereduction in bioaccessibility over time) of As(III) in thesesoils. The initial bioaccessibility of As(III) was significantlyhigher than the initial bioaccessibility of As(V) on thesame set of soils. However, over a 6-month period of aerobicaging, a significant portion of the solid-phase As(III) onthese soils was oxidized to As(V), decreasing its bioaccessibility markedly. A multivariable linear regression modelpreviously developed to predict the steady-state bioaccessibility of As(V) in soils was able to predict thebioaccessibility in As(III)-spiked soils within a root-mean-square error (RMSE) of 16.8%. Generally, soils having ahigher iron oxide content and lower soil pH exhibited lowerbioaccessibility. This model was also able to predict thein vivo bioavailability of As in contaminated soils previouslyused in an independent juvenile swine dosing trial withinan RMSE of 15.5%, providing a greatly improved yetconservative estimate of bioavailability relative to thetypical default assumption of 100%. However, the modelwas not able to accurately predict the bioavailability of Asin a different set of contaminated soils previously usedin an independent Cebus monkey dosing trial, consistentlyoverpredicting the bioavailability, resulting in an RMSEof 42.7%. This model can be used to provide an initialestimate of As bioavailability in soil to aid in screening sitesand justifying expensive site-specific animal feedingstudies. Further, as the model is based on major soilproperties, the resulting estimates are valid as long asthe major soil properties do not change, thus providing someconfidence in the long-term applicability of the estimates.
|
