The removal of dissolved zinc and copper from roof-runoff: A downpipe treatment system

Abstract

Anthropogenic activities related to urbanization and industrialization contribute high concentrations of heavy metals from urban stormwater runoff to waterways. In New Zealand, zinc (Zn) and copper (Cu) have been identified as the predominant heavy metals of concern because they have been observed to consistently exceed the Australian and New Zealand Environment and Conservation Council’s (ANZECC’s) guideline values for the protection of freshwater organisms in urban waterways. These heavy metals originate from a variety of sources, however, galvanized and copper roofs have been observed to contribute the highest per area Zn and Cu loads respectively. More so, >80% of the Zn and Cu released from these roofs are present in the dissolved reactive form making them more bioavailable and thus, potentially more toxic to aquatic organisms. Current stormwater management strategies have mostly focused on physical removal of particulate and particulate-bound contaminants while the dissolved contaminants are often left untreated. Also, it is difficult to retrofit conventional stormwater treatment devices such as retention ponds and raingardens in established urban areas due to limited space and the presence of underground services such as electricity, water and gas. Given that there are many existing Zn-and-Cu-based roofs in New Zealand whose runoff is discharged directly into the stormwater drainage system and/or waterways, there is a clear need to develop new at-source treatment devices that can remove these dissolved metals from roof-runoff. Sand have been the main treatment material used in stormwater filter systems, however, its removal efficiency for heavy metals have been observed to be low. As a result, this research explored the use of limestone, zeolite and mussel shells as treatment materials for the removal of dissolved Zn and Cu because of their high neutralization, adsorption and cation exchange capacity. The composition of calcium carbonate (CaCO3) in the structure of limestone and mussel shells, and the presence of alkali and alkaline earth elements sodium (Na+), potassium (K+), calcium (Ca2+) and magnesium (Mg2+) in the zeolite have neutralizing effects which causes the pH of the stormwater to increase, making it alkaline and consequently reducing the solubility of Zn and Cu. These three materials also possess high ion exchange capacity which allows them to adsorb large quantities of dissolved heavy metals. To evaluate the effectiveness of a downpipe treatment system containing limestone, zeolite and/or mussel shells in reducing the percentage of dissolved Zn and Cu from roof-runoff, both laboratory and field experiments were conducted. The laboratory experiments were done in two phases. Phase I consisted of batch experiments which assessed the dissolved Zn and Cu removal capacity of the three treatment materials at a grade of ≥ 1.18 ≤ 2.36 mm. Phase II was laboratory column treatment systems that were evaluated to quantify the hydraulic performance and dissolved Zn and Cu reduction capacity of each treatment material at two material depths (0.5 m and 1 m), two flow rates (1 L/min and 3 L/min) and when the materials were disturbed and undisturbed. For the laboratory column experiments, the untreated concentration of dissolved Zn and Cu from the roof-runoff ranged from 150-254 µg/L and 312-884 µg/L respectively. For the field experiments, Zn and Cu ranged from 406-2262 µg/L and 455-2581 µg/L respectively. The concentration of dissolved Zn and Cu in the untreated roof-runoff was considerably higher than ANZECC’s mixed instream guideline values for the protection of 90% of freshwater organisms of 15 µg/L and 1.18 µg/L for total Zn and Cu respectively. Evaluation of the percentage of dissolved Zn and Cu in the untreated roof-runoff from the laboratory column experiment showed that 100% of the Zn was in the dissolved form while dissolved Cu ranged from 78%-91%. These results indicate that Zn and Cu in roof-runoff is present mainly in the dissolved form which is ecotoxic to freshwater organisms. For the batch experiments, the percentage reduction of dissolved Zn and Cu varied. Limestone gave the highest mean percentage reduction for both Zn and Cu (87% Zn and 91% Cu) followed by mussel shells (78% Zn and 64% Cu) and then zeolite (48 % Zn and 64% Cu). However, for the laboratory column experiments, the amount of dissolved Zn and Cu removed by zeolite, limestone and mussel shells was not significantly different (p≤ 0.05). A reduction of 95-99% in dissolved Zn was achieved by all treatment materials at both depths, flow rates and disturbances while all three treatment materials only achieved 90-98% reduction in dissolved Cu at an undisturbed depth of 1 m. In the laboratory column experiments, all three treatment materials reduced dissolved Zn to concentrations well below ANZECC’s mixed instream guideline of 15 µg/L total Zn for the protection of 90% of freshwater organism’s. Although the reduction in dissolved Cu was not below ANZECC’s 90% mixed instream guideline of 1.8 µg/L total Cu, it was reduced to concentrations below 20 µg/L which was considerably lower than the 312 µg/L – 884 µg/L Cu present in the untreated roof-runoff. Dilution of the treated roof-runoff is expected as it moves downstream which would lead to further reduction in the concentration of dissolved Cu. The field experiment was conducted to collect data on the performance of the treatment system that would help improve the system design. Therefore, only mussel shells at an undisturbed depth of 1 m was evaluated. For the field experiment, dissolved Zn in runoff from the galvanized roof was reduced by 82-97% while dissolved Cu in runoff from the copper roof was reduced by 86-98%. These field results were comparable to what was obtained in the laboratory column experiments for mussel at an undisturbed depth of 1 m. These results show that the downpipe treatment system is robust and only small alterations to the system design would be required. From the laboratory column experiments, it was evident that adsorption and ion exchange was the main mechanism by which zeolite reduced the concentration of dissolved Zn and Cu. This is because there was no significant difference between the pH of the untreated roof-runoff and runoff treated by zeolite, however, dissolved Zn and Cu was reduced by >95% for the laboratory column experiments. It was also evident that neutralization contributed to the reduction of dissolved Zn and Cu in runoff treated by limestone and mussel shells in the batch experiments because a greater percentage reduction in dissolved Zn and Cu was observed at higher pH values. The use of zeolite, limestone and mussel shells were found to be very effective in removing dissolved Zn and Cu from roof-runoff in the laboratory column experiments. Mussel shell was then selected for field trials because it is a low cost and readily available waste product that proved to be just as effective as zeolite and limestone in the laboratory column experiments at removing dissolved Zn and Cu. The downpipe treatment system proved to be very effective under field conditions with >80% reduction in dissolved Zn and Cu achieved across each sampled rainfall event. These results show that factors such as the natural variation of untreated runoff quality that occurs under field conditions seem to have little influence on the Zn and Cu reduction efficiency of mussel shells which is an indication that the downpipe treatment system is effective and robust. Overall, this research contributes scientific understanding of a new stormwater treatment device that has the potential to achieve considerable reduction in dissolved Zn and Cu from roof-runoff that can be easily maintained and installed in urban areas with limited space.