Ultra- and microfiltration in drinking water production from surface water

The IWW Rheinisch-Westfaelisches Institut fuer Wasserforschung gemeinnuetzige GmbH, an Institute at the University of Duisburg, the Institute of Hygiene of the University of Bonn and the Wasserwerk des Kreises Aachen (WdKA) have realised a project supported by the German Federal Ministry for Education, Science, Research and Technology (BMBF). This project was to investigate whether membrane filtration (micro- or ultrafiltration) is able to enhance the particle removal efficiency of an existing drinking water treatment plant. Since the end of 1996 three pilot plants with a capacity of approx. 6-10 m(3)/h, one microfiltration (Memtec, Windsor, NSW 2756, Australia) and two ultrafiltration (Aquasource, 9 Avenue Alexandre Maistrasse, F 92500 Rueil Malmaison, France and X-Flow. BV, Bedrijvenpark Twente 289, 7602 KK Almelo, The Netherlands) plants, have been operated in dead-end mode with four different feed qualities: after the existing water treatment plant (flocculation, filtration stage, removal of iron and manganese by a second filtration stage); after flocculation, filtration stage and dosing of an oxidising agent; after flocculation, filtration stage, dosing of an oxidising agent and a CaCO3-filter; after the flocculation. The permeate of all pilot plants was impeccable except that Memtec microfiltration was not able to remove spiked viruses completely. Turbidity and particles >0.5 mu m were removed almost beneath the detection limit for all feed water qualities tested, it seems to be possible to replace the treatment step for the removal of manganese by ultrafiltration. However, in this case backwashing with only permeate or air is not able to remove the arising manganese layer and oxidising agents (NaOCl or H2O2) have to be added or disinfection backwashes have to be conducted, respectively. The adsorption of foulants - which usually fouled the membrane - at the manganese layer, which could be easily removed by the disinfection backwash, probably caused a higher performane of the ultrafiltration pilot plants. Furthermore, it was surprising that all pilot plants were able to treat water coming from the flocculation reaction basin of the existing treatment plant without passing the first filtration stage. In this case the X-Flow pilot plant has even reached better results, as in the case of operating after the existing treatment plant.