Water quality parameters affecting sodium silicate performance in partial lead service line replacements were examined using a fractional factorial experimental design and static pipe systems. An external copper wire was used to create a galvanic connection between a former lead service line and a new copper pipe. The pipe systems were filled with lab prepared water made to mimic real water quality. Water was changed on a three times per week basis. A 2(4-1) fractional factorial design was used to evaluate the impact of alkalinity (15 mg L(-1) or 250 mg L(-1) as CaCO3), nitrate (1 mg L(-1) or 7 mg L(-1) as N), natural organic matter (1 mg L(-1) or 7 mg L(-1) as dissolved organic carbon), and disinfectant type (1 mg L(-1) chlorine or 3 mg L(-1) monochloramine), resulting in eight treatment conditions. Fractional factorial analysis revealed that alkalinity, natural organic matter and monochloramine had a significant positive effect on galvanic current. Natural organic matter and monochloramine also had a significant positive effect with respect to both total and dissolved lead release. For the treatment conditions examined, 67-98% of the lead released through galvanic currents was stored as corrosion scales and predominantly comprised of particulate lead (96.1-99.9%) for all eight treatments. The use of monochloramine and the presence of natural organic matter (7 mg L(-1)) were not favourable for corrosion control in sodium silicate-treated partial lead service line replacements, although further studies would be required to characterize optimal water quality parameters for specific water quality types. For utilities operating with sodium silicate as a corrosion inhibitor, this work offers further evidence regarding the consideration of chlorine as a secondary disinfectant instead of monochloramine, as well as the value of controlling natural organic matter in distributed water.
