Pretreatment for desalination of seawater from an open intake by dual-media filtration: Pilot testing and comparison of two different media

The results of pilot testing of dual media filtration on seawater from an open intake in the Thermaikos Gulf, northern Greece, are presented in this study. Two different filter materials, FILTRALITE MC, a processed expanded clay material with particle range 1.5–2.5 mm, and anthracite coal with a similar particle range, were tested in parallel operation under various operating conditions (coagulant type and dose, filtration rate). Attention was focused on filtrate quality parameters relevant to feeding RO membrane desalination systems, that is, turbidity, silt density index (SDI), total organic carbon (TOC). In order to evaluate the effect of temperature variations on pretreatment performance, similar experiments were performed during a winter and a summer campaign of the same year. Despite the poor quality of feedwater, through an open intake, very low turbidities and SDI values lower than 5 could be obtained by means of coagulation and dual media filtration. Both filtration materials demonstrated almost similar performance in removing particulates from the feed water and produced permeates of acceptable quality for feeding RO systems (SDI15 values lower than 5) during the winter and summer campaigns. It was observed that both filters were much more effective during the summer period due to the higher temperatures prevailing, which allowed the formation of larger aggregates and their subsequent effective capture within the filter media. Laboratory and on site pilot tests showed that optimal performance was achieved when polyaluminum chloride (PACl) was added to the feed water as a coagulant medium at a dose 1.8 mg Al/L. In the winter period similar SDI15 values were obtained by both filters when operating at a flow velocity of 10 m/h, whereas at 5 m/h the sand/filtralite column produced water with lower SDI15 values and better quality compared to the one from the sand/anthracite column. On the other hand, during *Corresponding author. Presented at the conference on Desalination and the Environment. Sponsored by the European Desalination Society and Center for Research and Technology Hellas (CERTH), Sani Resort, Halkidiki, Greece, April 22–25, 2007. 0011-9164/06/$– See front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.desal.0000.00.000

A pilot unit for testing dual media filtration of raw seawater, under winter and summer conditions, was constructed and successfully operated within the facilities of TATE&LYLE, in Thessaloniki. Seawater, obtained through an open intake and used in the plant for cooling purposes, was made available for the study. Attention was focused on filtrate quality parameters relevant to feeding RO membrane desalination systems. A range of operating parameters such as flow velocities in the filters, as well as various coagulant types and concentrations were investigated in the laboratory and on site in order to identify conditions for optimal filter performance. It was demonstrated that acceptable filtrate quality for feeding RO systems was obtained by both types of media with the particular type of the raw feed water. It may be mentioned here that the origin of the feed water (being obtained through an open intake) is neither optimal for the design and operation of RO membrane pretreatment systems nor for membrane operation. Indeed, only the 2-min SDI of the particular feed water could be measured, which was very high (in the range of 38). Rough literature guidelines [5] suggest that media filtration alone cannot reduce this parameter by a factor much higher than 2. Thus, it may be argued that it is a significant achievement that very low turbidities and SDI values lower than 5 could be obtained in the present tests. Polyaluminum chloride (PACl) was found to be the most effective coagulant for this particular feed water. A concentration of 1.8 mg Al/L gave optimal results in terms of filtrate quality and acceptable duration of the filtration cycles. During the winter testing period the filtration cycles were determined by breakthrough of the retained solids in the columns and not by any significant buildup of hydrostatic head. The low quality of the feed water necessitated relatively short filtration cycles in the media filters. For this reason tests at flow velocities higher than 10 m/h were not possible during the winter campaign. In contrast, during the summer testing period, no solids breakthrough was observed. The higher feed water temperatures favoured the formation of large and strong aggregates, resulting in their effective capture in the filter bed. This improved efficiency of the coagulant allowed higher velocities (up to 15 m/h) to be tested. The filtration cycles, at all the flow velocities tested, were now determined by the available free space above filter materials for hydrostatic head buildup. Both filters, with FILTRALITE MC 1.5–2.5 mm or anthracite media, demonstrated a similar performance in removing particulates from the seawater and the reduction of the SDI15 value. Turbidities close to 0.1 NTU and SDI values lower than 4, or even close to 3, could be obtained. During the winter campaign, it was observed that at the higher flow velocity (10 m/h) the filtrate from the sand/filtralite column had almost the same quality as the one from the sand/anthracite column with respect to turbidity and SDI15 value. On the other hand, when operating at a lower flow velocity (5 m/h), the sand/ filtralite column achieved lower SDI15 values and filtrate water of better quality than the sand/ anthracite column. During the summer campaign both columns showed a similar behaviour in terms of filtrate quality, which was significantly improved compared to winter conditions. Finally, regarding the hydrostatic head development, it was observed that the sand/filtralite column had a slower head buildup than the sand/ anthracite column for the same duration of filter operation.