Brass plumbing components including meters, fittings and valves are used extensively in drinking water distribution systems. Until recently, most in-line brass components contained toxic lead, many of which are still presently in use. Corrosion of brass components leads to the release of metals to drinking water. The primary factors of brass corrosion in drinking water are temperature, alloy composition and water chemistry. In this thesis, a combination of mathematical modeling, analytical techniques and geochemical modeling were used to better understand what causes corrosion in brass components.
A comprehensive model for the release of copper, lead and zinc from brass water meters has been developed. This model provides a framework to evaluate how meter parameters, such as alloy composition and age, influence metal leaching from brass components. When considering brass composition, zinc concentration within the alloy is shown to be the primary factor in copper and zinc release. Brasses with greater than 8 – 9% zinc exhibit more rapid corrosion when compared to brasses with less than 8% zinc. Age was found to have more influence over lead release than alloy composition, with newer meters releasing significantly higher concentrations of lead versus older meters.
In addition to the oxidation of metallic surfaces, corrosion scale formation and dissolution also have a significant impact upon metal concentrations within drinking water. Optical microscopy, X-ray diffraction (XRD), Raman spectroscopy, and scanning electron microscopy (SEM) with energy dispersive X-ray spectroscopy (EDX), were used to characterize the morphology and mineralogy of corrosion scale in two sets of residential water meters. The meters, which were in service for up to 40 years, came from two locations with contrasting water chemistries; Seattle, with relatively low alkalinity, hardness and total dissolved solids (TDS), and Cincinnati, which has moderate alkalinity, hardness and TDS. Results showed the copper minerals cuprite and malachite, to be most abundant within the corrosion scale from both sources. Lead minerals were much more prevalent within the Cincinnati meters, as were carbonates (both Cu & Pb). In general, the Cincinnati meters contained more substantial and consistent scale coverage whereas coverage on the Seattle meters was patchier and more localized.
Real world use of drinking water systems cycle between flow and periods of stagnation, where water sits quiescently within the system. During stagnation, changes in water chemistry can include metal concentration, solution pH, and oxidation reduction (redox) potential. PHREEQC was utilized to calculate the saturation index (SI) of metallic species with changing water chemistry. The SI values were used to evaluate whether a given mineral will dissolve or precipitate from water during stagnation. These values were compared to the mineralogy identified in the meters to better understand the mechanism of scale development. Changes in redox potential had the most significant effect upon SI values. Minerals present within the scale were found to form under distinctly different conditions suggesting that dissolution and precipitation rates must also be considered.