Intermittent slow sand filters: improving their design for developing world applications

Date
2010
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
As many as 500,000 people in developing countries rely on an intermittent slow sand filter, called the BioSand filter to provide safe drinking water. However, it has been shown that BioSand filters do not remove all of the pathogens. The mechanisms of water purification in slow sand filtration have not been clearly understood until this day. In order to successfully influence intermittent slow sand filters performance we need to have a full understanding of removal processes that govern their efficiency. In the first part of this study pathogen removal mechanisms in intermittent slow sand filtration were investigated. In the second part of this study the effect on pathogen removal of a zero-valent iron-sand mixed layer, placed at the bottom of an intermittent slow sand filter was investigated. A series of experiments was performed on sand columns ripened with creek water. It was shown that the majority of E.coli was removed during filtration through the schmutzdecke, while the residence time in the biologically active layer did not have a significant effect on its removal. Quite to the contrary, the majority of MS-2 virus was removed with the residence time in the biologically active layer while the filtration through the schmutzdecke did not have significant affect on its removal. Clearly, two different mechanisms govern bacteria and virus removal in intermittent slow sand filtration. The EDX analysis of the schmutzdecke material revealed that it was primarily composed of O, C, Si, Al, Fe, Ca, and Mn. The zeta potential analysis showed that the schmutzdecke was neutral in charge and was composed of positively, neutrally and negatively charged particles. It was determined that a flocculation process was not a likely mechanism of either E.coli or MS-2 removal in intermittent slow sand filtration. The disturbance of the schmutzdecke had a significant impact on E.coli removal but only a slight effect on MS-2 removal. The intricate internal structure of the schmutzdecke might affect bacteria removal by an increase in rate of collisions. The disturbance of the schmutzdecke may cause the water to bypass the treatment in the schmutzdecke “network” of polymeric strands. It was concluded that an additional treatment step is needed in order to efficiently remove viruses; that the residence time in the schmutzdecke affects the quality of the effluent.; that the disturbance of the schmutzdecke might lead to a significant compromise in water quality and thus should be avoided; and that the addition of Al or Fe into the filter influent might benefit the pathogen removal efficiency. In the second part of the study an intermittent slow sand filter was amended with a zero-valent iron-sand mixture at the bottom of the column in order to improve its performance, especially virus removal. In the first challenge experiment the ZVI amended filter removed 100% of both E.coli and MS-2. Although, in the second challenge experiment the zero-valent iron-sand layer caused a decrease in E.coli/MS-2 concentration for unknown reasons, both E.coli and MS-2, exhibited breakthrough in the effluent. The ZVI powder can be obtained by sieving iron filings which are a waste byproduct of iron industry. Therefore, it is readily available to communities in developing countries. It was shown that use of ZVI in an intermittent slow sand filter can greatly improve its safety. However, the mechanisms responsible for this benefit and limitations of ZVI are not yet explained and deserve to be investigated.
Description
Keywords
Citation