Applicability of using sedimentation and membrane filtration for stormwater treatment

Stormwater runoff is one of the most important contributor of pollutants to natural waters. Therefore, separation of pollutants from stormwater runoff before entering the receiving waters is an important issue. Nitrogen (ammonium, nitrite, nitrate, total nitrogen), phosphorus (organic and inorganic), metals (Fe, Cd, Zn, Pb, As, Al, metalloid arsenic, Cr, Cu, Ni in total, dissolved and truly dissolved form) are main compounds that are contained in various concentrations in the runoff and should be separated. Various methods have been used for stormwater treatment. Membrane technology has shown acceptable treatment ability in water and wastewater industries1, but is not yet used for treatment of stormwater. Ceramic, polymeric and metallic membranes have been used in other research studies in order to optimise removal efficiency for separation of particulate, colloidal and dissolved substances in runoff. To prevent fouling, which is the major problem in membranes and to decrease the operational cost, use of a proper pre-treatment is helpful 2–4, as is periodical cleaning of membrane surface 5,6. This process can be further assisted by using membranes with larger pore size (microfiltration, ultrafiltration) than nanofiltration and reverse osmosis membranes which have smaller pore size but higher removal efficiency. In this study, sedimentation was used as a pre-treatment method and a membrane process with different pore sizes and pulsatile flow (fluid flow with periodical variation) were evaluated with respect to separation of pollutants from a snowmelt mixture. For the aim of this research work, polyether sulfone nanofiltration and ultrafiltration membranes (PES NF, UF membrane) were used to assess the ability of the overall process to postpone fouling and treat the snowmelt mixture. The variation of pulse frequency was chosen between 0 and 2 Hz. A peristaltic pump with a constant rotational velocity of 220 rpm was used in all the experiments. Sedimentation as pre-treatment method was investigated in pre-experiments and it was shown that a variation in sedimentation time between 0.5-1 hours did not affect the treatment ability of the process. Therefore, 0.5 hour of sedimentation time was chosen in order to reduce the experimental run time. The experimental runs will be carried out during the winter period Feb-April 2021. The concentration of TSS, pH, total and dissolved metals and chloride, as well as the particle size distribution, turbidity and volume of clean water achieved in each experimental run is measured. The results from these experiments will give valuable information about treatment efficiency, if pulsatile flow increases the treatment performance as well as knowledge about the clogging process of the filter