The 4th International Conference on Advances in Civil and Ecological Engineering Research (ACEER 2022)

Abstract: Arsenic (As) is carcinogenic and a widespread groundwater contaminant throughout the world. Zerovalent iron (ZVI) can reduce As contamination via adsorption/co-precipitation with Fe oxides formed during its corrosion. Applying ZVI in the field has often occurred in a trial-and-error manner, i.e. by monitoring contaminant concentrations downstream the groundwater flow followed by additional injections in many cases. The mechanisms and the immobilization capacity towards As of different types of ZVI at complex, realistic, and comparable field conditions are poorly known. We studied an As-contaminated aquifer in Hjältevad, Sweden, with As concentration at ca. 1,400 µg.L-1 in the groundwater and ca. 50 mg.kg-1 in dried sediments. Anoxic conditions throughout the aquifer is consistent with the predominance of As(III) species, which are more harmful and mobile.
Four ZVI sorbents were tested in the laboratory under oxygen-free atmosphere. They have different particle sizes and different coatings: micro and nano ZVI, with and without sulfidic shell. Experiments were conducted in two complementary ways: batch experiments and columns experiments.
All sorbents decreased the As concentration in solution. However, parameters such as pH, redox potential, and the rate at which As was removed suggests that immobilization kinetics and mechanisms may be different. Before ZVI addition, As K-edge XANES shows that As(V) is dominant in the solid phase whereas As(III) dominates the aqueous phase. The sorption of aqueous As(III) onto the newly formed Fe phases takes place via the oxidation of As(III) to As(V), despite the reductive conditions created by the addition of Fe(0), confirmed by the sharp decrease in redox potential. Moreover, Fe K-edge EXAFS indicates that different Fe corrosion products, formed at different rates, coexist depending on particle size and coating.
As a result, the As K-edge XANES fingerprint differs depending on the type of ZVI used, suggesting that different Fe corrosion products lead to different mechanisms of As uptake. Finally, the residual As that remained in solution after ZVI treatment shifted from mostly As(III) in the untreated groundwater to As(V) after ZVI addition.
Our results are important to elucidate the pathways of As sorption by different ZVI materials at realistic conditions in anoxic aquifers. This should provide more reliable predictions on the long-term strength of As immobilization in order to maximize often costly and time-consuming field efforts.