| The EPA has promulgated a more stringent drinking water standard for arsenic, reducing the MCL from 50 mug/L to 10 mug/L. One of the most promising technologies for selective arsenic removal from ground water sources is adsorption using metal oxide adsorbents. A major problem faced by utilities considering adsorption media is that laboratory or field scale pilot testing of these media requires very long runs and therefore is very costly. Developed for GAC adsorption, the rapid small-scale column test is an accelerated testing method that can be used to determine sorption characteristics by measuring the performance of small diameter media at higher hydraulic loading rates. Results from RSSCTs can be scaled up to predict the performance of full-scale treatment systems. It is hypothesized that this method will be applicable for arsenic adsorption onto metal oxide media.; This project began with a series of batch experiments to determine the adsorption characteristics of four arsenic adsorption media; an iron oxide coated activated alumina (AA), granular ferric hydroxide (GFH), another granular ferric oxide (SORB E-33), and a proprietary media (SANS) developed by Sandia National Laboratories. Batch isotherm results were fitted to the Freundlich isotherm to determine parameters K and 1/n, which were crucial in later modeling efforts.; The second phase of the investigation consisted of extensive column testing using three sizes of each media. Preliminary results showed a linear relationship between the surface diffusion coefficient and particle size; therefore, a linear relationship was assumed between empty bed contact time and particle size for RSSCT scaling purposes. For exact similarity between small- and large-scale columns, the breakthrough curves for a particular media of different size should be identical. Breakthrough curves obtained, although similar, were not identical with a variation in particle size. This led to the conclusion that the surface diffusion coefficients were not linearly proportional to particle size and the linear scaling relationship was inaccurate. Upon determination of an accurate relationship between surface diffusivity and particle size, the scaling ratios were adjusted as appropriate. Results indicate that RSSCTs can be used to simulate the performance of full-scale arsenic adsorption columns. |
