Mercury Removal and Reduction by Pumice Supported Iron Nanoparticles from Contaminated Groundwater

Abstract

Iron nanoparticles (nZVI) are extensively used for the remediation of organic and inorganic pollutants from contaminated water. Supporting materials like kaolinite, chitosan, and pumice are used to increase the dispersion and stability of nZVI. This study focused on removal of Hg from contaminated groundwater, where Hg concentration is comparatively lower than wastewater and have comparatively higher nitrate and NOM concentration with varying dissolved oxygen (DO) levels. Pumice supported nZVI (p-nZVI) was synthesized using 0.5 - 1 mm pumice as supporting material and iron nanoparticles in synthesized PnZVI had an average diameter of 95.4 nm. Hg removal and reduction efficiency was tested at various Fe(0) fractions in p-nZVI, p-nZVI loadings, and Hg loadings. It was found that p-nZVI with 7.7% iron content has highest Hg removal efficiency (> 96 %) for 250 nM of Hg(II) spiked in the solution. An increase in Hg dose, decreased the removal efficiency and increase in Fe(0) fraction in PnZVI increased removal efficiency. Freundlich isotherm was better fit showing the heterogeneous nature of p-nZVI surface. For further insight of DO and nitrate effect batch experiments were conducted in anoxic and oxic conditions. It was found that volatilized mercury (Hg0) and Fe(II) were higher in anoxic conditions, whereas lower soluble Hg was found in solution phase in oxic environment. The addition of nitrate showed an increase in headspace Hg(0) while Hg removal efficiency was similar to without nitrate suspensions. The mercury removal efficiency was more than 95% in oxic environments in all cases while it didn’t increase more than 87% in anoxic environments. Oxic environment lead to the production of ferric iron in extensive amount by Fe(0) corrosion present in p-nZVI, that resulted in more adsorption sites for Hg. To see natural organic matter (NOM) effect on Hg reduction and removal by p-nZVI Suwannee river natural organic matter (SRNOM), glutathione (GSH), and 9,10-Anthraquinone-2,6-disulfonic acid disodium salt (AQDS) were tested in oxic conditions. SRNOM has inhibition effect on Hg removal and reduction efficiency by p-nZVI, whereas GSH decreased the reduction efficiency and increased removal efficiency. Surprisingly AQDS didn’t show any effect on either reduction or removal of Hg by p-nZVI. Lepidocrocite was mainly found as secondary mineral in suspensions with GSH, AQDS and without NOM that provides adsorption sites for Hg. SRNOM adsorbed on the reactive surface of nZVI and decreased its reactivity towards Hg reduction and removal. GSH complexation with Hg resulted in lower headspace Hg(0). Lastly p-nZVI filled columns were tested for in situ Hg removal from groundwater at flow rates of 0.5 and 2.0 L per day with 500 nM of Hg spiked in artificial groundwater for bench scale column. In 223 days of continuous flow Hg removal efficiency for the columns was more than 99 % and no breakthrough was observed. Miniature column tested with flowrate of 3.4 L/d and 5 mg/L of influent Hg showed a break through time of 40 hours and removed 174 mg of total Hg per gram of Fe upto break through time. It shows that p-nZVI is a promising material for permeable reactive barriers and column application for in situ groundwater treatment.