1、OPTIMIZING COAGULATION AND DIRECT FILTRATION PROCESSESFOR LOW TURBIDITY, LOW TEMPERATURE WATERSKEYWORDSCoagulation, Velocity Gradients, Low turbidity Water,Natural Organic matter (NOM), Direct FiltrationABSTRACTThis study aims at exploring optimum coagulant dose and optimum pH for a low turbidity, l
2、ow temperature water. Coagulation of low temperature and low turbidity waters generally requires addition of clays for adequate quantities of hydro-oxide precipitate to remove colloidal matter. This study focuses on particulate and natural organic matter removal from low temperature and low turbidit
3、y water without the addition of clays. An optimum combined dose of ferric sulfate and a cationic polyelectrolyte at an optimum pH using pilot scale direct filtration train was achieved. This work, not only recommended the elimination of a sedimentation basin but substantially reduced chemical cost f
4、or coagulation and pH adjustment. It was observed that use of color removal instead of turbidity removal as a variable of direct filtration performance was well suited to low turbidity waters.INTRODUCIONAll surface waters that are used for water supply require treatment to remove contaminants both f
5、or health as well as aesthetic reasons. Major contaminants that are found in surface waters are clays, microorganisms, natural organic matter (NOM) and trace metals in a few cases. Solid-liquid separation processes such as coagulation and filtration are often employed to remove these undesirable con
6、taminants. These processes when optimized, can remove all organic, inorganic and suspended matter to a level below national water quality standards.Low turbidity waters are hard to coagulate due to low concentrations of stable particles. Before the discovery of Giardia problem, many treatment plants
7、 in US used to stop adding coagulant during winter and spring months as the turbidity of the incoming water was already below the national primary drinking water standards. A few treatment plants would add only token coagulants to, presumably, take advantage of filtration. Filtration, though in plac
8、e, was effective only after effective coagulation Al-Jadhai, 1993.Color in water stems from decayed vegetation, humic substances and natural as well as anthropogenic organic sources. Effective color removal is an indication of effective coagulation. Lefebvre et. al.2003 reported that NOM can be best
9、 removed at pH around 6 while using Ferric Iron as sole coagulant.On one hand color in water can serve as surrogate for NOM Edzwald, 1979, on the other hand only effective coagulation would result in effective filtration as transport and attachment are the dominant steps in both operations Habibian
10、et. al. 1975. Whilst turbidity represents the bulk water characteristics and cannot provide any insight on gradation of colloidal and suspended matter, removal of microscopic pathogens of certain size range cannot be ensured. Particle counting and particle size distribution (PSD) reveal a clear pict
11、ure about the size distribution of particles that have been removed or reduced. One disadvantage of particle counting is the coagulation of submicron particles into microscopic zone. Rise in total number of particles may not correlate with decrease in turbidity. This study was conducted to optimize
12、pH and coagulant dose using a pilot scale direct filtration plant in parallel with an existing conventional treatment plant in Rhode Island USA. Stakeholders sought to explore cost reduction measures including changes in the treatment train, if economical. Rather than traditional variable of effluen
13、t turbidity, color and particle counts were used to determine treatment effectiveness.Major objectives of the study were to determine the optimum dose of ferric sulfate with and without polymer addition using color and particle counting as performance indicators, and, to examine the effectiveness of
14、 direct filtration scheme compared with conventional treatment train. This study covers a comparison between the pilot plant and the full scale water treatment plant for low turbidity water for winter months only. No data was collected or compared for the summer time. Water temperature in Rhode Isla
15、nd varies between -10 to + 20 Co. No data was collected for biological variables.MATERIALS AND METHODSExperimental investigations of this study were divided into three phases: Phase-1: Laboratory scale investigations involving optimum coagulant dose and pH determination. Phase-2: Pilot scale confirm
16、ation of the effectiveness of the optimum dose at optimum pH. Phase-3: Comparison of pilot plant results with actual plant.Twenty-five jar testing experiments were conducted in phase-1 within a pH range of 6-10 and ferric sulfate dose of 0-20 mg/L. Color, remaining after coagulation and settling, was used as the indicator of performance.Polymer was added as secondary coagulant in the second phase of jar testing experiments. A total of twenty-s
