Conditions Contributing to Poor Coagulation and Sedimentation Performance in Municipal Water Treatment

Abstract

The failure of the Auburn Water Works municipal water treatment coagulation and sedimentation process causing the plant to shut down from June 29, 1999 to July 2, 1999 was investigated. The investigation proceeded in two phases. The first phase was a series of jar tests consisting of pure, deionized water that was spiked with a combination of substances to induce specific desired test parameters common to municipal waters before treatment. The purpose of this phase of testing was to ascertain the variables that have the most pronounced influence on coagulation and sedimentation behavior after treatment with aluminum sulfate (alum, Al2SO4); an effort to reveal the key variables that contribute to coagulation failure. The second phase of testing was to take the worst case combination of variables from phase I and treat them with a matrix of other coagulant chemicals. In this way, it may be determined which chemical, if any, would produce better treatment results than alum. A matrix of jar testing for each of the two primary variables (pH and alkalinity) was examined for all of the coagulants. This second phase will focus primarily on determining proper processing conditions. Other coagulants used were ferric chloride (FeCl3), cationic polymer, anionic polymer, non-ionic polymer, and poly-aluminum chloride. Results of the investigation demonstrated that treating a water at pH 8 yields better results than a pH of 6. Although some coagulants lower the pH in water to a pH closer to 6, those waters yielded better turbidity and color test results when the pH was raised back up to 8 than did waters left at the lower pH. It was also revealed through the test results that the waters with higher alkalinities treated at pH 8 yielded more favorable results than did lower alkaline samples at the same pH. Also, the investigation showed that alum is the best coagulant for municipal water treatment based on turbidity and color testing, but ferric chloride yielded acceptable results as well. Results produced with all of the polymeric coagulants except poly-aluminum chloride proved far less desirable than did alum and ferric chloride results and in some cases had little to no effect whatsoever on the final water quality. Poly-aluminum chlorides on the other hand did have some possible benefit over alum in terms of dose requirements and range of pH during coagulation.