Arsenic removal by zero-valent iron: field, laboratory and modeling studies

• Published in: Water research (Oxford) Vol. 37; no. 6; pp. 1417 - 1425
• Main Authors: Nikolaidis, Nikolaos P, Dobbs, Gregory M, Lackovic, Jeffrey A
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01-03-2003
• Subjects: Arsenic; Arsenic - chemistry; Arsenic - isolation & purification; Environmental Monitoring; Filtration; Iron - chemistry; Modeling; Models, Theoretical; Soil Pollutants - isolation & purification; Water Pollutants - isolation & purification; Water treatment technology; Zero valent iron; Arsenic; Water treatment technology; Modeling; Zero valent iron
• ISSN: 0043-1354; 1879-2448
• DOI: 10.1016/S0043-1354(02)00483-9

Field and laboratory studies were conducted to elucidate the design factors and mechanisms of arsenic removal from contaminated ground water using zero-valent iron. Large scale, field pilot experiments demonstrated for more than 8 months that iron filing filters can efficiently remove arsenite from aqueous solutions to levels less than 10
μg/L. The maximum arsenic accumulation measured was 4.4
mg As/g of media. The iron filing filters leached significant quantities of iron (73% of the iron was leached). A critical design parameter of the system was found to be the hydraulic detention time of the water in the filter. TCLP analyses of the spent media indicated that the arsenic concentration in the leachate was two orders of magnitude lower than the 5
mg/L of TCLP for arsenic. Spectroscopic and laboratory arsenic leaching studies (alkaline extraction and TCLP) suggest that the arsenic surface precipitate is related to sulfur. The aging process (due to the longevity of the removal mechanism) makes the precipitation process virtually irreversible. A mathematical model was developed to simulate the removal process using a partitioning coefficient and a mass transfer process. Calibration of these parameters using the data for three columns revealed that the equilibrium-partitioning coefficient was the same for all three columns while the mass transfer coefficient was a function of the flow rate. The calibrated mass transfer coefficients are similar to those reported in the literature if they are normalized to the surface area of the media.