A laboratory study of copper and zinc runoff during first flush and steady-state conditions

The influence of environmental conditions and corrosion layer characteristics have been investigated on the runoff rate of copper and zinc, used as roofing material. For this purpose, a rain device has been constructed, capable of simulating rain episodes of varying intensity and pH, and used on new and aged copper and zinc panels of varying origin and corrosion product composition. The setup, using artificial rain with a composition resembling the southern and central part of Sweden, has proven to result in realistic runoff rates for all materials investigated. During a rain event, easily soluble corrosion products will be removed in the first rain volume, commonly referred to as the first flush, followed by a more or less constant runoff rate during subsequent rain. The magnitude of the first flush depends on environmental conditions prior Co a rain episode, e.g., length of dry periods and extent of dry deposition, as well as on rain volume and rain intensity. A defect-rich and porous corrosion layer increases the magnitude of the first flush. The total metal runoff quantity increases with decreasing pH for both copper and zinc. In agreement with previous findings outdoors, an effect of patina age can be seen on copper. The laboratory data can be used to explain variations in runoff rate between different sampling periods observed in field data. (C) 2000 Elsevier Science Ltd. All rights reserved.