Abstract
Consumption of drinking water containing arsenic at concentrations even below the World Health Organization provisional limit of 10 μg/L can still lead to unacceptable health risks. Consequently, the drinking water sector in the Netherlands has recently agreed to target 1 μg/L of arsenic in treated water. Unfortunately, in many poor, arsenic-affected countries, the costs and complexity of current methods that can achieve <1 μg/L are prohibitive, which highlights the need for innovative methods that can remove arsenic to <1 μg/L without costly support infrastructure and complicated supply chains. In this work, we used Fe(0) electrolysis, a low cost and scalable technology that is also known as Fe(0) electrocoagulation (EC), to achieve <1 μg/L residual dissolved arsenic. We compared the arsenic removal performance of green rust (GR), ferric (oxyhydr)oxides (Fe(III) oxides) and magnetite (Mag) generated by EC at different pH (7.5 and 9) in the presence of As(III) or As(V) (initial concentrations of 200–11,000 μg/L). Although GR and Fe(III) oxides removed up to 99% of initial arsenic, neither Fe phase could reliably meet the 1 μg/L target at both pH values. In contrast, EC-generated Mag consistently achieved <1 μg/L, regardless of the initial As(V) concentration and pH. Only solutions with initial As(III) concentrations ≥2200 μg/L resulted in residual arsenic >1 μg/L. As K-edge X-ray absorption spectroscopy showed that Mag also sorbed arsenic in a unique mode, consistent with partial arsenic incorporation near the particle surface. This sorption mode contrasts with the binuclear, corner sharing surface complex for GR and Fe(III) oxides, which could explain the difference in arsenic removal efficiency among the three Fe phases. Our results suggest that EC-generated Mag is an attractive method for achieving <1 μg/L particularly in decentralized water treatment.
Original language | English |
---|---|
Article number | 115170 |
Number of pages | 11 |
Journal | Water Research |
Volume | 168 |
DOIs | |
Publication status | Published - 1 Jan 2020 |
Funding
We gratefully acknowledge funding provided by the Dutch Organization for Scientific Research in a Veni Grant to CMvG (Project No. 14400 ). We thank Marcel Ceccato for assistance with FI-HG-AAS measurements. Synchrotron experiments were performed partly at the DUBBLE beamline at the ESRF, Grenoble, France, with assistance from Dipanjan Banerjee. We also thank Ryan Davis and Sharon Bone for technical support during synchrotron data collection at SSRL. Use of SSRL, SLAC National Accelerator Laboratory, was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences , under Contract No. DE-AC02-76SF00515 . Appendix A
Keywords
- Arsenic treatment
- Electrocoagulation
- EXAFS spectroscopy
- Mineral formation
- Oxyanion incorporation