Influence of in vitro assay pH and extractant composition on As bioaccessibility in contaminated soils

In vitro bioaccessibility assays are often utilised to determine the potential human exposure to soil contaminants through soil ingestion. Comparative studies have identified inconsistencies in the results obtained with different in vitro assays. In this study we investigated the potential causes for the variability between in vitro assay results using the PBET and SBRC assays to assess As bioaccessibility in 5 brownfield contaminated soils. Total As concentration in the 5 soils ranged from 227 to 807 mg As kg(-1) in the <250 μm particle fraction while XANES analysis identified that As was predominately present as As(V) (>88% was sorbed to Fe mineral phases) with the remaining As present as beudantite or orpiment mineral phases. Arsenic bioaccessibility varied depending on the in vitro methodology; markedly higher values were obtained using the SBRC gastric phase compared to the PBET gastric phase, however, similar As bioaccessibility values were obtained in both the SBRC and PBET intestinal phases. The difference in As bioaccessibility following SBRC and PBET gastric phase extraction appeared to be due to the difference in gastric phase pH (i.e. 1.5 versus 2.5 respectively), however, modifying the PBET gastric phase to pH1.5 (that of the SBRC gastric phase) enhanced As bioaccessibility up to 1.6 fold, but was still markedly lower than SBRC values. Although As bioaccessibility was enhanced, the increase did not occur as a result of the solubilisation of As associated Fe mineral phases suggesting As bioaccessibility may also be strongly influenced by the in vitro assay extractant composition. The extractant composition of the PBET assay incorporates a number of organic acids in addition to pepsin which may inhibit the sorption of As onto iron oxide surfaces, therefore increasing As solubility at the modified (pH1.5) gastric phase pH.