In 1990, the town of Hopkinton, Mass., instituted a polyphosphate sequestrant feed in two of its five wells to address colored water complaints caused by source water iron and manganese. Failure to meet the Lead and Copper Rule (LCR) action levels and continuing intermittent red water problems prompted the water utility to seek an alternative treatment strategy. After analyses of the water quality of the different wells, the scales on some lead service lines, the operational patterns, the location of lead service lines, and the distribution system configuration, a combination of sodium silicate (with chlorination) and sodium hydroxide addition at different wells was investigated. At study monitoring sites, an initial silicate dose of 25–30 mg/L elevated the pH from 6.3 to 7.1 and immediately resulted in a 55% reduction in lead levels and an 87% reduction in copper levels. An increase to a silicate dose of 45–55 mg/L elevated the pH to 7.5 and produced greater reductions in lead and copper. The treatment change reduced 90th percentile lead and copper levels by at least 95%, enabling compliance with the LCR. The aesthetic quality of the drinking water after treatment was equal or superior to the quality before treatment. The study showed that for many small and medium water systems with multiple wells and entry points, simultaneous overall corrosion control and sequestration of iron and manganese is possible, and it avoids the adverse drinking water/wastewater consequences associated with polyphosphate addition.

To assess the interactions between pipe materials, organic carbon levels, and disinfectants, studies using annular reactors with ductile-iron, polyvinyl chloride (PVC), epoxy, and cementlined coupons were carried out in the laboratory and at four field sites. Laboratory studies used biologically treated water with and without 0.2 mg/L residual free chlorine or monochloramine, in the presence or absence of 0.5 or 2.0 mg/L humic substances. In the lab studies, the type of disinfectant did not lead to significant differences in effluent and biofilm counts. Increases in carbon led to greater numbers of biofilm and effluent organisms, with the effect most pronounced on iron. Regardless of carbon level, PVC systems typically had the lowest numbers of bacteria, whereas iron had the highest. Cement and epoxy were intermediate. Depending on the site, field studies showed that iron had the highest number of bacteria or there was no difference in materials.olving bacterial regrowth problems (e.g., positive coliforms) has historically been linked to improving source water treatment or optimizing secondary disinfection in the distribution system. Specifically, most water systems have battled their bacterial regrowth problems by increasing disinfectant residual, converting from free chlorine (Cl 2 ) to chloramine, or installing source water treatment processes to decrease the concentration and/or change the nature of the organic matter entering the distribution system. All of these approaches rely on modifying water quality parameters. Another possible solution may be in the rehabilitation or replacement of distribution system infrastructure.A growing body of data strongly indicate that the amount of attached bacteria on iron pipe can be several orders of magnitude higher than on other materials, such as cement linings, polyvinyl chlorinde (PVC), or polyethylene (Niquette et al

More and more utilities are using chloramines in place of free chlorine for greater residual stability and better compliance with both the Total Coliform Rule and the stringent requirements of the Disinfectants/Disinfection Byproducts Rule. However, new information about disinfectantinduced changes in oxidation reduction potential, lead and copper chemistry, scale formation, and scale destabilization involving natural organic matter has contributed to greater understanding of factors influencing increased corrosion and metals release. This article consolidates and updates information about potential effects of changing disinfectants on lead and copper release in drinking water distribution systems. The findings indicate the importance of chemical properties and electrochemical behavior in understanding how corroding metals and alloys respond to transition from free chlorine to chloramines and vice versa.

In laboratory, pilot, and Lead and Copper Rule (LCR) monitoring studies of copper solubility, cupric hydroxide or a highly microcrystalline tenorite appears to be the dominant solid phase, supporting the theoretical cupric hydroxide copper solubility model. This model predicts that as pipe ages, tenorite or malachite will form and predominate as the surface phase, lowering copper levels. Orthophosphate treatment initially may lower copper levels and yield temporally stable copper concentrations instead of displaying this aging phenomenon. Given enough time (i.e., years to decades), systems without orthophosphate likely will experience copper concentrations lower than what would be achieved with orthophosphate. Because LCR‐compliance monitoring structure is intentionally biased toward sites that may exhibit elevated lead rather than elevated copper, water utilities should be aware that copper concentrations above the action level and several times the 90th percentile value can be prevalent in parts of their distribution systems. To achieve maximum health protection and strict regulatory compliance, systems with higher percentages of newer construction with copper water lines may need to consider the tradeoffs between pH and/or alkalinity adjustment and orthophosphate addition for their optimal corrosion‐control treatment strategy.

Evaluation of treatment alternatives for corrosion by means of pipe loop systems requires a viable statistical basis for decision‐making. Decision‐makers evaluating corrosion treatment alternatives under the Lead and Copper Rule need the statistical tools to choose the optimum treatment from pipe loop studies. In many cases, the data generated by pipe loop studies do not follow a normal distribution; therefore, nonparametric statistics apply. Techniques that are discussed include determination of data normality using the Kolmogorov–Smirnov and chisquare tests, determination of stabilization using the Spearman coefficient, and comparison of treatments using the Wilcoxon signed ranks or rank sum test. Other data issues that are important to evaluating corrosion studies include determining sample size and frequency, determining the confidence and accuracy of results, and evaluating data outliers.