Electrocoagulation Pretreatment Prior to Ultrafiltration
Open Access
- Author:
- Timmes, Thomas Charles
- Graduate Program:
- Environmental Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- March 04, 2009
- Committee Members:
- Brian Dempsey, Dissertation Advisor/Co-Advisor
Brian Dempsey, Committee Chair/Co-Chair
Fred Scott Cannon, Committee Member
Jack Vincent Matson, Committee Member
Herschel Adams Elliott, Committee Member
Steven Robert Richards, Committee Member - Keywords:
- Electrocoagulation
Pretreatment
Ultrafiltration
Membrane
Fractionation - Abstract:
- The objective of this research was to determine the feasibility of using electrocoagulation (EC) as a pretreatment prior to ultrafiltration (UF). Wastewater, seawater, and natural waters were investigated as source waters. A continuous flow bench-sale EC reactor was configured, calibrated, and operated under constant current conditions to deliver aluminum or iron coagulant doses. The efficiency of the in-line EC pretreatment step, the UF membrane performance, and the UF permeate quality were compared to pretreatment results obtained using an equivalent molar dose of aluminum or iron from a chemical coagulant (aluminum sulfate or ferric chloride, respectively). A novel, neutral-pH natural organic material (NOM) fractionation system provided insight as to the effectiveness of the electrocoagulation pretreatment to target the hydrophobic/hydrophilic (HPO/HPI) acid fraction of organic matter suspected to be the dominant foulant of ultrafilters. For pretreatment of wastewater effluent organic materials (EfOM), in-line EC (aluminum) pretreatment better removed the HPO/HPI acid fraction of EfOM under all coagulation conditions. For this wastewater effluent, EC pretreatment was better for removal of total organic carbon (TOC) and UV254 adsorbance, while alum pretreatment resulted in a comparatively lower rate of transmembrane pressure (TMP) increase and better flux recoveries after both hydraulic and chemical cleaning. At sub-critical fluxes, variations in the rate of TMP increase between the comparative EC and alum pretreatment conditions are slight and essentially equivalent. For pretreatment of modified Atlantic Ocean seawater, optimal doses from EC (iron) and ferric chloride provided statistically similar reductions in TMP and hydraulic resistance during constant sub-critical flux ultrafiltration compared to the baseline condition (no coagulant). At super-critical fluxes, the EC generated cake layer remained comparatively more porous (and less compressible) under increasing TMP. It was also more efficiently removed by hydraulic and chemical cleaning than the ferric chloride-based cake layer, which resulted in higher pure water flux recoveries. In related pilot-scale studies using Pacific Ocean seawater, EC (iron) pretreatment resulted in slightly higher TMP than observed following ferric chloride pretreatment during long-term sub-critical flux ultrafiltration. In this study, substantial accumulation of precipitates developed on the iron electrodes and were identified by X-ray diffraction as magnetite, maghemite, lepidocrocite, and akaganeite phases. For pretreatment of a surface water mixture, an underdosing condition using either dosing platform increased the TMP and hydraulic resistance during ultrafiltration and resulted in poorer flux recoveries after hydraulic and chemical cleanings. Pretreatment doses resulting in a visible floc (despite a low zeta potential and streaming current) lowered TMP and hydraulic resistance for all flux conditions. At this dose, EC (iron) pretreatment offered improved flux recovery following hydraulic cleaning, but ferric chloride pretreatment achieved a greater flux recovery after chemical cleaning. These findings were confirmed during longer-term constant sub-critical flux ultrafiltration using polysulfone hollow-fibers. Anoxic conditions (dissolved oxygen <0.5 mg/L) during in-line iron electrocoagulation and ultrafiltration yielded poor flux recovery after chemical cleaning, but adverse effects were not observed during sub- and super-critical flux experiments. This research indicates that EC offers a potential feasible and effective pretreatment strategy for mobile drinking water production facilities.