Alexander R. Maginnis (maginna@email.uc.edu)

Paul L. Bishop, (Paul.Bishop@UC.edu)

Darren A. Lytle (Lytle.Darren@epamail.epa.gov)

In this study, microelectrodes were utilized in order to measure levels of DO (dissolved oxygen) and pH as a function of distance from the wall of a copper drinking water pipe and as a function of time at a fixed distance (50mm) from the pipe wall. Such data can be used to better model drinking water systems, to better understand and reduce red water formation, and to provide additional insight into corrosion as it relates to many scientific and industrial applications. Characterizing microenvironments at or near the solid-liquid interface within drinking water pipes is an integral part of understanding the complete oxidation reaction taking place between water and a reactive metal surface. This kind of high spatial resolution measurement has been made possible relatively recently by the onset and advancement of microelectrode technology. Microelectrodes with tip diameters of 4-7 mm were constructed for this experiment.

An 8" length of 1.5" diameter new copper pipe, with a hole cut in the side for access, is rinsed in acetone and acid washed and soaked in order to remove any residues left during the manufacturing or storage of the pipe. The openings are then sealed, and the pipe section is conditioned using a laboratory prepared water sample. Prepared water samples consist of MilliQ water of either a DIC (dissolved inorganic carbon concentration) of 2 mg C/L or 100 mg C/L (added as NaHCO₃) and a pH of either 8.0 or 6.5 (adjusted by addition of HCl or NaOH). The pipe is conditioned for at least two days by dumping and refilling and oxygenating at the same time each day, with the last filling occurring immediately before the experiment, in order to begin the experiment under a saturated DO content.

Microelectrode measurements as a function of distance were first recorded. Distance profiles starting at the wall and moving toward the surface varied significantly from measurements starting in the bulk solutions and moving back toward the wall during the same experiment. This may be a result of the rapid change in water chemistry over the time period of the experiment. In order to further investigate this variance over time, distance has been held constant at 50 mm from the wall throughout subsequent experiments, with measurements taken periodically over an 8-hour period of time. Bulk pH and temperature measurements were also recorded throughout the experiment, and bulk DO measurements were recorded before and after the experiment whenever possible. Results to date indicate a faster drop in DO at an initial pH of 6.5 than at an initial pH of 8.0. For conditions of pH = 6.5 and DIC = 2 mg/L, DO concentrations have been shown to level out at about 5 mg O₂/L after about 3.5 hours. Conversely, at a pH of 8.0 and a DIC of 100 mg C/L, the depletion of oxygen is much slower, showing slowly declining DO levels still near saturation at a corresponding time of 3.5 hours. Additionally, plots of DO and pH taken at a DIC of 100 mg C/L are more linear, exhibiting more stable trends, than at a lower a DIC of 2 mg C/L which has a correspondingly lower buffer capacity. These findings support hypotheses based on theoretical and empirical chemical kinetics.

Additional experiments investigate this phenomena using an excavated and conditioned piece of 4" diameter iron pipe which exhibits scaling, heavy tuberculation, well developed corrosion cells, and pitting. These experiments utilize DO and pH distance profiles, under more stable conditions than the aforementioned copper pipe data.