Critique of Jar Testing for the Design of Coagulation-Flocculation Systems

Abstract

Abstract The effects of clay type and concentration, jar shape, paddle shape, rapid mix time and flocculation velocity gradient on batch, bench-scale jar tests of turbidity were studied using a fractional factorial design. The two-parameter kinetic model for flocculation of Agraman, Kaufman and Bratby was used. This model provided a good fit of the data when they were log-transformed. The variables that had a statistically significant effect on the kinetics constants for floc agglomeration and breakup were the interaction of clay concentration and flocculation velocity gradient, plus clay type combined with rapid-mix time. The jar shape and the paddle shapes did not have significant effects. The kinetic constants measured in this analysis agreed with those reported previously by Agraman and Kaufman. The work assumes that turbidity is an acceptable measure of the number concentration. The results confirm the usefulness of the kinetic model and emphasize the importance of testing the goodness-of-fit of the model before analyzing the data.