No description available.

The ban on lead plumbing materials in the Safe Drinking Water Act (1986) and the EPA Lead and Copper Rule (1991) have successfully reduced lead contamination of potable water supplies. The success of these regulations gave rise to a belief that serious lead contamination was an important past problem that had been solved, and that additional fundamental research was therefore unnecessary. This work carefully re-examined the lead contamination issue from the perspective of 1) new regulations causing a shift from chlorine to chloramine disinfectant, 2) assumptions guiding sampling strategies, 3) existing performance standards for brass, and 4) galvanically driven corrosion of lead bearing plumbing materials. The results were instrumental in uncovering and understanding a serious problem with lead contamination in Washington, D.C. A critical reading of the literature indicates that chloramines can accelerate corrosion of lead bearing materials and increase lead contamination of water. When a new sampling protocol was conceived and used in Washington homes to assess the nature of the problem, hazardous levels of lead were found to be present in some drinking water samples. Contrary to the conventional wisdom, lead was not always highest in first draw samples, but often increased with flushing. This has several important implications for monitoring and public health. For instance, well-intentioned public education materials were causing consumers to drink water containing very high levels of lead in some circumstances. Laboratory and field-testing proved that chloramines were causing serious lead corrosion problems. That testing also discovered that, unbeknownst to scientists and utilities, free chlorine itself can act as a corrosion inhibitor, reducing lead solubility and contamination of water. The net result is that changing disinfectant from free chlorine to chloramine can sometimes trigger serious lead contamination of water. While the worst problems with lead in Washington, D.C. came from the lead services, significant levels of lead were occasionally sampled from homes with solders or brass as the lead source. This prompted re-evaluation of the ANSI/NSF 61, Section 8 standard, which is relied on to protect public health from in-line brass plumbing devices that might leach excessive lead to potable water. In-depth study of the standard revealed serious flaws arising from use of a phosphate buffer in the test waters and a failure to control carbonate dissolution from the atmosphere. Due to these deficiencies, small devices made of pure lead could actually pass the performance test. The public therefore has no assurance that devices passing NSF Section 8 testing are safe and reforms to the standard are obviously needed. Other problems arise from connecting copper pipe to lead bearing plumbing in practice. The copper is cathodic and dramatically accelerates corrosion of the lead anode via a galvanic current. Corrosion and hydrolysis of released Pb²⁺ can lower pH near the surface of the lead and increase its solubility. A similar galvanic effect can arise from cupric ions present in the water via deposition corrosion mechanism. In cases where part of a lead service line is replaced by copper pipe, the galvanic corrosion effect can create a serious long-term problem with lead contamination. Such partial lead service line replacements are occurring in many US cities and the practice should be stopped. Lead contamination of potable water is not only a problem of the past but also of the present. While additional research is necessary before regulators, utilities and homeowners can anticipate and mitigate such problems with confidence, this work provides sound fundamental basis for future progress. / Master of Science

A study was conducted to investigate and quantify the effects of corrosion inhibitors on metal release within a pilot distribution system while varying the source water. The pilot distribution system consisted of pre-existing facilities from Taylor et al (2005). Iron, copper, and lead release data were collected during four separate phases of operation. Each phase was characterized by the particular blend ratios used during the study. A blended source water represented a water that had been derived from a consistent proportion of three different source waters. These source waters included (1) surface water treated through enhanced coagulation/sedimentation/filtration, (2) conventionally treated groundwater, and (3) finished surface water treated using reverse osmosis membranes. The corrosion inhibitors used during the study were blended orthophosphate (BOP), orthophosphate (OP), zinc orthophosphate (ZOP), and sodium silicate (Si). This document was intended to cite the findings from the study associated with corrosion treatment using various doses of sodium silicate. The doses were maintained to 3, 6, and 12 mg/L as SiO2 above the blend-dependent background silica concentration. Sources of iron release within the pilot distribution system consisted of, in the following order of entry, (1) lined cast iron, (2) un-lined cast iron, and (3) galvanized steel. Iron release data from these materials was not collected for each individual iron source. Instead, iron release data represented the measurement of iron upon exposure to the pilot distribution system in general. There was little evidence to suggest that iron release was affected by sodium silicate. Statistical modeling of iron release suggested that iron release could be described by the water quality parameters of alkalinity, chlorides, and pH. The R2 statistic implied that the model could account for only 36% of the total variation within the iron release data set (i.e. R2 = 0.36). The model implies that increases in alkalinity and pH would be expected to decrease iron release on average, while an increase in chlorides would increase iron release. The surface composition of cast iron and galvanized steel coupons were analyzed using X-ray photoelectron spectroscopy (XPS). The surface analysis located binding energies consistent with Fe2O3, Fe3O4, and FeOOH for both cast iron and galvanized steel. Elemental scans detected the presence of silicon as amorphous silica; however, there was no significant difference between scans of coupons treated with sodium silicate and coupons simply exposed to the blended source water. The predominant form of zinc found on the galvanized steel coupons was ZnO. Thermodynamic modeling of the galvanized steel system suggested that zinc release was more appropriately described by Zn5(CO3)2(OH)6. The analysis of the copper release data set suggested that treatment with sodium silicate decreased copper release during the study. On average the low, medium, and high doses decreased copper release, when compared to the original blend source water prior to sodium silicate addition, by approximately 20%, 30%, and 50%, respectively. Statistical modeling found that alkalinity, chlorides, pH, and sodium silicate dose were significant variables (R2 = 0.68). The coefficients of the model implied that increases in pH and sodium silicate dose decreased copper release, while increases in alkalinity and chlorides increased copper release. XPS for copper coupons suggested that the scale composition consisted of Cu2O, CuO, and Cu(OH)2 for both the coupons treated with sodium silicate and those exposed to the blended source water. Analysis of the silicon elemental scan detected amorphous silica on 3/5 copper coupons exposed to sodium silicate. Silicon was not detected on any of the 8 control coupons. This suggested that sodium silicate inhibitor varied the surface composition of the copper scale. The XPS results seemed to be validated by the visual differences of the copper coupons exposed to sodium silicate. Copper coupons treated with sodium silicate developed a blue-green scale, while control coupons were reddish-brown. Thermodynamic modeling was unsuccessful in identifying a controlling solid that consisted of a silicate-based cupric solid. Lead release was generally decreased when treated with sodium silicate. Many of the observations were recorded below the detection limit (1 ppb as Pb) of the instrument used to measure the lead concentration of the samples during the study. The frequency of observations below the detection limit tended to increase as the dose of sodium silicate increased. An accurate quantification of the effect of sodium silicate was complicated by the observations recorded below detection limit. If the lead concentration of a sample was below detection limit, then the observation was recorded as 1 ppb. Statistical modeling suggested that temperature, alkalinity, chlorides, pH, and sodium silicate dose were important variables associated with lead release (R2 = 0.60). The exponents of the non-linear model implied that an increase in temperature, alkalinity, and chlorides increased lead release, while an increase in pH and sodium silicate dose were associated with a decrease in lead release. XPS surface characterization of lead coupons indicated the presence of PbO, PbO2, PbCO3, and Pb3(OH)2(CO3)2. XPS also found evidence of silicate scale formation. Thermodynamic modeling did not support the possibility of a silicate-based lead controlling solid. A solubility model assuming Pb3(OH)2(CO3)2 as the controlling solid was used to evaluate lead release data from samples in which lead coupons were incubated for long stagnation times. This thermodynamic model seemed to similarly describe the lead release of samples treated with sodium silicate and samples exposed to the blended source water. The pH of each sample was similar, thus sodium silicate, rather than the corresponding increase in pH, would appear to be responsible if a difference had been observed. During the overall study, the effects of BOP, OP, ZOP, and Si corrosion inhibitors were described by empirical models. Statistically, the model represented the expected value, or mean average, function. If these models are to be used to predict a dose for copper release, then the relationship between the expected value function and the 90th percentile must be approximated. The USEPA Lead and Copper Rule (LCR) regulates total copper release at an action level of 1.3 mg/L. This action level represents a 90th percentile rather than a mean average. Evaluation of the complete copper release data set suggested that the standard deviation was proportional to the mean average of a particular treatment. This relationship was estimated using a linear model. It was found that most of the copper data sub-sets (represented by a given phase, inhibitor, and dose) could be described by a normal distribution. The information obtained from the standard deviation analysis and the normality assumption validated the use of a z-score to relate the empirical models to the estimated 90th percentile observations. Since an analysis of the normality and variance (essentially contains the same information as the standard deviation) are required to assess the assumptions associated with an ANOVA, an ANOVA was performed to directly compare the effects of the inhibitors and corresponding doses. The findings suggested that phosphate-based inhibitors were consistently more effective than sodium silicate when comparing the same treatment levels (i.e. doses). Among the phosphate-based inhibitors, the effectiveness of each respective treatment level was inconsistent (i.e. there was no clear indication that any one phosphate-based inhibitor was more effective than the other). As the doses increased for each inhibitor, the results generally suggested that there was a corresponding tendency for copper release to decrease.

This study evaluated the effects of orthophosphate (OP) inhibitor addition on iron, copper, and lead corrosion on coupons exposed to different blends of groundwater, surface water, and desalinated seawater. The effectiveness of OP inhibitor addition on iron, copper, and lead release was analyzed by statistical comparison between OP treated and untreated pilot distribution systems (PDS). Four different doses of OP inhibitor, ranging from zero (control) to 2 mg/L as P, were investigated and non-linear empirical models were developed to predict iron, copper, and lead release from the water quality and OP doses. Surface characterization evaluations were conducted using X-ray Photoelectron Spectroscopy (XPS) analyses for each iron, galvanized steel, copper, and lead/tin coupon tested. Also, a theoretical thermodynamic model was developed and used to validate the controlling solid phases determined by XPS. A comparison of the effects of phosphate-based corrosion inhibitor addition on iron, copper, and lead release from the PDSs exposed to the different blends was also conducted. Three phosphate-based corrosion inhibitors were employed; blended orthophosphate (BOP), orthophosphate (OP), and zinc orthophosphate (ZOP). Non-linear empirical models were developed to predict iron, copper, and lead release from each PDS treated with different doses of inhibitor ranging from zero (control) to 2 mg/L as P. The predictive models were developed using water quality parameters as well as the inhibitor dose. Using these empirical models, simulation of the water quality of different blends with varying alkalinity and pH were used to compare the inhibitors performance for remaining in compliance for iron, copper and lead release. OP inhibitor addition was found to offer limited improvement of iron release for the OP dosages evaluated for the water blends evaluated compared to pH adjustment alone. Empirical models showed increased total phosphorus, pH, and alkalinity reduced iron release while increased silica, chloride, sulfate, and temperature contributed to iron release. Thermodynamic modeling suggested that FePO4 is the controlling solid that forms on iron and galvanized steel surfaces, regardless of blend, when OP inhibitor is added for corrosion control. While FePO4 does not offer much control of the iron release from the cast iron surfaces, it does offer protection of the galvanized steel surfaces reducing zinc release. OP inhibitor addition was found to reduce copper release for the OP dosages evaluated for the water blends evaluated compared to pH adjustment alone. Empirical models showed increases in total phosphorus, silica, and pH reduced copper release while increased alkalinity and chloride contributed to copper release. Thermodynamic modeling suggested that Cu3(PO4)2·2H2O is the controlling solid that forms on copper surfaces, regardless of blend, when OP inhibitor is added for corrosion control. OP inhibitor addition was found to reduce lead release for the OP dosages evaluated for the water blends evaluated compared to pH adjustment alone. Empirical models showed increased total phosphorus and pH reduced lead release while increased alkalinity, chloride, and temperature contributed to lead release. Thermodynamic modeling suggested that hydroxypyromorphite is the controlling solid that forms on lead surfaces, regardless of blend, when OP inhibitor is added for corrosion control. The comparison of phosphate-based inhibitors found increasing pH to reduce iron, copper, and lead metal release, while increasing alkalinity was shown to reduce iron release but increase copper and lead release. The ZOP inhibitor was not predicted by the empirical models to perform as well as BOP and OP at the low dose of 0.5 mg/L as P for iron control, and the OP inhibitor was not predicted to perform as well as BOP and ZOP at the low dose of 0.5 mg/L as P for lead control. The three inhibitors evaluated performed similarly for copper control. Therefore, BOP inhibitor showed the lowest metal release at the low dose of 0.5 mg/L as P for control of iron, copper, and lead corrosion.

The effects of corrosion inhibitors on water quality and the distribution system were studied. This dissertation investigates the effect of inhibitors on iron surface roughness, copper surface roughness, and copper release. Corrosion inhibitors included blended poly/ortho phosphate, sodium orthophosphate, zinc orthophosphate, and sodium silicate. These inhibitors were added to a blend of surface water, groundwater, and desalinated brackish water. Surface roughness of galvanized iron, unlined cast iron, lined cast iron, and polyvinyl chloride was measured using pipe coupons exposed for three months. Roughness of each pipe coupon was measured with an optical surface profiler before and after exposure to inhibitors. For most materials, inhibitor did not have a significant effect on surface roughness; instead, the most significant factor determining the final surface roughness was the initial surface roughness. Coupons with low initial surface roughness tended to have an increase in surface roughness during exposure, and vice versa, implying that surface roughness tended to regress towards an average or equilibrium value. For unlined cast iron, increased alkalinity and increased temperature tended to correspond with increases in surface roughness. Unlined cast iron coupons receiving phosphate inhibitors were more likely to have a significant change in surface roughness, suggesting that phosphate inhibitors affect stability of iron pipe scales. Similar roughness data collected with new copper coupons showed that elevated orthophosphate, alkalinity, and temperature were all factors associated with increased copper surface roughness. The greatest increases in surface roughness were observed with copper coupons receiving phosphate inhibitors. Smaller increases were observed with copper coupons receiving silicate inhibitor or no inhibitor. With phosphate inhibitors, elevated temperature and alkalinity were associated with larger increases in surface roughness and blue-green copper (II) scales.. Otherwise a compact, dull red copper (I) scale was observed. These data suggest that phosphate inhibitor addition corresponds with changes in surface morphology, and surface composition, including the oxidation state of copper solids. The effects of corrosion inhibitors on copper surface chemistry and cuprosolvency were investigated. Most copper scales had X-ray photoelectron spectroscopy binding energies consistent with a mixture of Cu2O, CuO, Cu(OH)2, and other copper (II) salts. Orthophosphate and silica were detected on copper surfaces exposed to each inhibitor. All phosphate and silicate inhibitors reduced copper release relative to the no inhibitor treatments, keeping total copper below the 1.3 mg/L MCLG for all water quality blends. All three kinds of phosphate inhibitors, when added at 1 mg/L as P, corresponded with a 60% reduction in copper release relative to the no inhibitor control. On average, this percent reduction was consistent across varying water quality conditions in all four phases. Similarly when silicate inhibitor was added at 6 mg/L as SiO2, this corresponded with a 25-40% reduction in copper release relative to the no inhibitor control. Hence, on average, for the given inhibitors and doses, phosphate inhibitors provided more predictable control of copper release across changing water quality conditions. A plot of cupric ion concentration versus orthophosphate concentration showed a decrease in copper release consistent with mechanistic control by either cupric phosphate solubility or a diffusion limiting phosphate film. Thermodynamic models were developed to identify feasible controlling solids. For the no inhibitor treatment, Cu(OH)2 provided the closest prediction of copper release. With phosphate inhibitors both Cu(OH)2 and Cu(PO4)·2H2O models provided plausible predictions. Similarly, with silicate inhibitor, the Cu(OH)2 and CuSiO3·H2O models provided plausible predictions.