Key parameters controlling arsenic dynamics in coastal sediments: An analytical and modeling approach

The coupling of analytical tools and modeling approaches allowed the identification of the key parameters controlling arsenic (As) dynamics in anoxic marine sediments. Eh, pH, dissolved/particulate As, Fe, Mn, S, P, and organic/inorganic C were determined in sedimentary porewaters and in the sediments itself at 3 geochemically contrasting stations and at different seasons, in the heavily contaminated, semi-closed Toulon bay (NW Mediterranean Sea). Elemental analysis and selective extractions allowed the identification of As distribution in the studied sediments. Organic matter (OM) quality was characterized by fluorescence measurements. A 1D steady-state modeling (PROFILE) was used to estimate depth reaction intervals and reaction rates by fitting the measured element profiles. Thermodynamic simulation (PHREEQC) was also performed to calculate the chemical speciation and simulate the dissolved/particulate As fractionation. This work demonstrated the consistency of the additive speciation/adsorption modeling to simulate As profiles in marine porewaters while considering the different sedimentary phases and the species that could potentially act as competitors (e.g. CO32 −, PO43 −, …). By taking into account the presence of OM and appropriately adjusting its reactivity toward As, the simulated dissolved As profiles consistently fitted the measured one. This As–OM reactivity is related to the organic matter quality, determined by fluorescence measurements. samples, As dynamics in the subsurface sediments was shown to be strongly linked to the iron cycling, especially to amorphous iron oxyhydroxide, through diagenesis reactions. The Fe/As ratio involved in the diagenesis processes (determined both experimentally and from the PHREEQC simulation of the dissolved As profiles without organic matter interaction) was close to ~ 210. Higher ratio was predicted by the simulation when considering the As–organic matter interaction. A similar observation was found for other mineral phases in sulfidic sediments (sulfide, clays …). The decreased affinity of As for inorganic minerals could be due to a competition between organic/inorganic phases or to a ternary arsenic–OM–inorganic phase association, as recently suggested in the literature. Trials provided OM–As stability constant values in these samples, ranging (in log10) from 2 to 3.2 for particulate OM and from 2 to 4.7 for dissolved OM. Therefore, organic matter was clearly shown to play a role in As dynamics in sediments or porewater (and thus potentially in other natural samples), even in the presence of inorganic phases. OM, thus, should be added in databases and taken into account in future simulations.