Retention of Iodide by Bacteriogenic Iron Oxides

Abstract

This study was performed to determine the ability of wetland bacteriogenic iron oxides (BIOS) to immobilize iodide in contaminated groundwater systems near Chalk River, Canada. The sorption of iodide onto synthetic hydrous ferric oxide (HFO) and BIOS was investigated using an autotitrator and an I⁻ ion-selective electrode to generate high-resolution anion sorption data over a pH range of 2.5 to 9. The effect of strontium sorption in the presence of I⁻ was also investigated to determine its effect on iodide retention as it is also a common contaminant near Chalk River. Both HFO and BIOS correspond to 2-line ferrihydrite with surface areas of 227.7 m² g⁻¹ and 92.52 m² g⁻¹, respectively. Sorption of I⁻ was found to be pH dependent for both HFO and BIOS and was most strongly immobilized at pH 2.5. The pH at which 50% of the I⁻ was bound to HFO occurred at pH 4.0, whereas BIOS maintained 50% sorption to pH 9. Field data also indicated a 54% decrease for iodine and 75% for ¹²⁹I in waters passing over in-situ BIOS at circumneutral pH. Iodide sorption to HFO is best explained by homogeneous functional groups, whereas sorption of I⁻ to BIOS is best explained by heterogeneous functional groups, due to the presence of bacterial functional groups with pKa values that extend to 9.0. The presence of Sr²⁺ decreased iodide sorption on HFO by 10–20%, but had no effect on BIOS due to its surface functional groups being reactive over the pH range investigated in this study. These results imply that BIOS is a useful sorbent for natural retention of I⁻ from groundwater and that the amount of organic material present in iron oxides is an important factor when considering remediation strategies for radionuclides in groundwater.

Keywords:

• iodide
• strontium
• sorption
• iron oxides
• bacteria
• groundwater
• nuclear waste

ACKNOWLEDGMENTS

This work was supported by an NSERC Strategic Grant to F. Grant Ferris. The authors thank two anonymous reviewers for their helpful suggestions with this manuscript. We also thank the Ogilvies for their continued assistance with field sampling, Sean Langley for sample collection and processing, Chris Weisener for BET analysis and Alex Ibrahim for his assistance in preparing Figure 1.