Investigation and Optimization of Dual Coagulation ProcessesEPA Grant Number: R822462
Title: Investigation and Optimization of Dual Coagulation Processes
Investigators: Benjamin, Mark M. , Edwards, Marc
Institution: University of Washington , University of Colorado at Boulder
EPA Project Officer: Lasat, Mitch
Project Period: October 1, 1995 through September 30, 1998 (Extended to April 30, 1999)
Project Amount: $273,209
RFA: Exploratory Research - Engineering (1995) Recipients Lists
Research Category: Engineering and Environmental Chemistry , Land and Waste Management
Description:The goal of this project is to improve the understanding of coagulation processes designed to meet goals of good coagulation of particles and removal of natural organic matter (NOM). Specific objectives of this projects are: (1) to evaluate the trade-offs between doses of conventional coagulants and polymers with respect to meeting the treatment objectives; (2) to characterize the NOM removal by conventional coagulants; and (3) to correlate the observations made in meeting the first two objectives with properties of the sludges formed. The results of the project will potentially be applicable in the establishment of conceptual models and practical guidelines for 'dual coagulation' (coagulation using both polymers and metal salts), allowing water utilities to reduce usage of conventional coagulants while achieving satisfactory particulate and NOM removal.
The project is being carried out by two research groups. Work at the University of Washington focuses on the role of polymers in dual coagulation. The objectives of this part of work are to determine the effect of polymer addition on the treatability of flocs, to evaluate the impact of polymer addition on NOM removal, and to establish an optimization procedure for coagulant dosing using dual coagulation. Preliminary results indicate that most of the achievable NOM removal can often be accomplished at much lower coagulant doses than are necessary for optimal particulate coagulation. Since the addition of polymer (coagulant aid) presumably improves the treatability of flocs, it might be possible to achieve satisfactory removal of particles and NOM while reducing sludge generation significantly by using a low dose of metal coagulant along with a small amount of polymer. However, since both flocculation and NOM removal depend on solution pH, coagulant dose, and DOC concentration of the water, optimization of the operation parameters can be very complicated. An operational diagram which synthesizes the relevant information, i.e., which illustrates the effects of operational parameters on both NOM removal and flocculation, is currently under development. With the use of this site-specific diagram, a water utility can determine the optimal operational conditions for specified treatment criteria and estimate the sensitivity of the process to the operating conditions and influent water quality.
Work at the University of Colorado has focused on development of a generalized model to predict the efficiency with which dissolved NOM can be insolubilized by addition of iron- or aluminum-based coagulants. This effort synthesizes a large database of coagulation experiments conducted around the country into a comprehensive model that can serve to guide water treatment plant operation and regulatory decision-making. The model uses as inputs only parameters that are routinely measured by researchers and practitioners dealing with coagulation and NOM removal processes. The basic conceptualization assumes that all DOC can be divided into two distinct fractions: one fraction that forms strong complexes with metal hydroxide surfaces formed during coagulation and another that does not. The distribution of molecules within each fraction and their response to coagulant addition are assumed to be identical in all water sources; however, the fraction of DOC that is in a given fraction is assumed to vary among water sources. Also, the affinity of the hydroxide for sorbable NOM is assumed to depend only on the type of coagulant and the final pH. Preliminary results show that these assumptions are statistically valid, and the final residual DOC in the coagulated water at specified operation conditions (e.g., solution pH, coagulant type and dose, and original water quality) can be accurately predicted by the proposed model.