AWWA ACE58167

This work demonstrates that aluminum residuals in drinking water can increase susceptibility to copper pitting corrosion under certain conditions of pH and chlorine residual. In the first set of experiments examining the role of pH, free chlorine was increased incrementally every two days from zero to 4.8 mg/L as Cl2 using bleach (6% NaOCl). The target level of chlorine was increased as follows: 0, 0.5, 0.8, 1.2, 1.5, 1.8, 2.1, 2.4, 3.6, and 4.8 mg/L as Cl2. Targeted pH values of 6,7,8, and 9 were tested in duplicate (8 cells total), and the pH was adjusted and Cl2 residuals were replenished twice daily. pH values were continuously held to within 0.3 units of the target value. In all experiments, free chlorine was measured using the DPD colorimetric test per standard method 4500-Cl G. The water was completely changed on day 12 and tests were ended on day 20 of the experiment. Cumulative copper release to the water was measured at these times using the bicinchoninate method. Corrosion potential (Ecorr) and polarization resistance (RP) measurements were collected hourly using a multiplexer connected to a personal computer, using a polarization scan rate of 0.5 mV/sec from -5 to +5 mV of the open circuit potential. Platinum counter and Ag-AgCl reference electrodes were checked and maintained regularly to insure accuracy. Instantaneous corrosion rates were estimated using potentiodynamic scans and Tafel analyses. The combined impacts of Al and Cl2 were examined by repeating the above tests, but with addition of 2 mg/L as Al of Al(OH)3 solids. The Al(OH)3 stock solution was created by raising the pH of 3000 mg/L as Al-AlCl3 solution to 8.0 using NaOH, allowing the solids to settle, decanting the water, re-suspending the solids in freshwater, until the conductivity of the suspension was equal to that of the experimental water. The same chlorine dosing schedule was used, except an additional level of 6.0 mg/L Cl2 was added on day 20 for all pH conditions. Copper release was measured and water was changed on day 22, and chlorine was again dosed to 6.0 mg/L. The pH for all conditions was then raised to 9.0, and Ecorr and Rp were measured for another 2 days. To account for possible deposition of aluminum solids within the experimental apparatus, aluminum solids were re-dosed once approximately 24 hours after each water change. Thus, the actual level of aluminum in the water could have been as much as 50% higher than the target levels, and lower than the target levels dependent on losses to the apparatus surface or pipe samples. At the end of the experiments, samples were cut in half and photographs were taken of the corroded copper surfaces using a digital camera. Includes 7 references, figures.