Date of Award

Spring 1-1-2015

Document Type


Degree Name

Master of Engineering (ME)

First Advisor

R. Scott Summers

Second Advisor

JoAnn Silverstein

Third Advisor

Fernando F. L. Rosario-Ortiz


Coagulation is a commonly used treatment practice for drinking water applications. Traditionally, coagulation has been used for particle removal. However, with the emergence of disinfection byproduct (DBP) regulations, which are formed through a reaction between dissolved organic matter (DOM) and the disinfectant chlorine, coagulation is also used to remove of DOM, to inhibit the formation of DBPs. Research has been published exploring the kinetics of particle formation during coagulation, but there is little to no published research examining DOM coagulation kinetics.

Three treatment approaches that are commonly carried out around the coagulation process and may be impacted by DOM coagulation kinetics are: (1) the use of powdered activated carbon (PAC) to remove seasonal taste and odor (T&O) compounds [2-methylisoborgeol (MIB) or geosmin], (2) direct filtration where conventional sedimentation is omitted, and (3) disinfection with chlorine. These treatment processes could be further optimized if DOM coagulation kinetics were known.

DOM kinetic coagulation experiments were performed on eight surface waters with a range of water quality parameters. It was found that DOM coagulation occurs within the first ten to twenty seconds of the coagulation process. Temperature, pH, rapid mix mixing speed, turbidity, coagulant type, and coagulant dose had no effect on the kinetics of coagulation in the range evaluated.

Use of PAC, direct filtration, and chlorination were examined after DOM coagulation had occurred. PAC was less effective after DOM coagulation, while chlorination formed fewer DBPs when applied after DOM coagulation. Direct filtration demonstrated little to no carryover of DOM through filtration once DOM coagulation occurred.

PAC addition was further investigated as traditionally PAC was added at the rapid mixing stage. PAC was added at four different contact times prior to coagulation, simultaneously with a coagulant, and at two different addition times after coagulation. It was found that PAC worked most effectively in removing MIB when applied prior to coagulation. Earlier addition times of PAC provide more removal of MIB, but would also require large reactors with associated capital costs. The ten-minute addition time prior to coagulation was chosen as the most efficient location for PAC addition to remove MIB.