Author's personal copy Journal of Membrane Science 297 (2007) 141–151 Controlling submicron particle deposition in a side-stream membrane bioreactor: A theoretical hydrodynamic modelling approach incorporating energy consumption Tao Jiang a,b,∗ , Maria D. Kennedy b , ChangKyoo Yoo c , Ingmar Nopens a , Walter van der Meer d , Harry Futselaar e , Jan C. Schippers b , Peter A. Vanrolleghem a,f a BIOMATH, Ghent University, Coupure Links 653, B-9000 Gent, Belgium b UNESCO-IHE for Water Education, P.O. Box 3015, 2601 DA Delft, the Netherlands c College of Environmental and Applied Chemistry/Center for Environmental Studies, Kyung Hee University, Yongin-city, Gyeonggi-do 446-701, South Korea d Vitens Fryslˆ an, Snekertrekweg 61, Postbus 400, 8901 BE Leeuwarden, the Netherlands e Norit Process Technology B.V., P.O. Box 741, 7500 AS ENSCHEDE, the Netherlands f modelEAU, D´ epartement de g´ enie civil, Pavillon Pouliot, Universit´ e Laval, Qu´ ebec, QC, G1K 7P4, Canada Received 27 July 2005; received in revised form 13 March 2007; accepted 18 March 2007 Available online 23 March 2007 Abstract Soluble microbial products (SMP) in the sludge water phase are regarded as the main foulant in MBRs. This study further developed an existing hydrodynamic model by incorporating energy consumption. The focus was on the cost-effectiveness of crossflow (CF) velocity in the control of submicron particle deposition. A sensitivity analysis showed that CF had the greatest impact on both particle backtransport and energy consumption. The other operational variables, i.e., dry solid content (DS), membrane tube dimension (D and L) and temperature (T) were generally less influential with respect to particle backtransport and energy consumption. Submicron particles were likely to deposit in side-stream MBRs, and the lowest backtransport velocity was found for particle radii around 0.1 m and CF below 0.5 m/s. A particle size distribution (PSD) profile of MBR sludge showed a main peak at 40 m and a second peak at 0.1–1 m. The abundance of submicron particles at 2000 kDa was confirmed by a Liquid chromatography–Organic Carbon Detection (LC-OCD) analysis. The colloids responsible for the second peak in the PSD received high weighting factors (high filter cake formation potential) in the model optimization. In a lab-scale MBR, this critical crossflow velocity was between 0.75 and 1 m/s at 40 L/(m 2 h). © 2007 Elsevier B.V. All rights reserved. Keywords: Membrane bioreactor; Fouling; Hydrodynamics; Modelling; Optimization 1. Introduction Membrane bioreactors (MBRs) are an innovative activated sludge process using membrane filtration instead of secondary clarifiers to achieve biomass separation. The microfiltration or Abbreviations: AS, absolute sensitivity; BW, backwashing; CF, cross- flow; DS, dry solids; EPS, extracellular polymeric substances; LC-OCD, liquid chromatography–organic carbon detection; NFR, normalized fouling rate; PSD, particle size distribution; RS, relative sensitivity; SEC, size exclusion chro- matography; SMP, soluble microbial products; TMP, transmembrane pressure; WWTP, wastewater treatment plant ∗ Corresponding author at: BIOMATH, Ghent University, Coupure Links 653, B-9000 Gent, Belgium. Tel.: +32 9 264 59 35; fax: +32 9 264 62 20. E-mail address: tao.jiang@biomath.ugent.be (T. Jiang). ultrafiltration membrane produces excellent effluent quality free of particulates and coliforms, which is suitable for many reuse applications [1]. The high sludge concentration and the elimina- tion of secondary clarifiers save space, which makes the MBR an attractive option for space limited situations (e.g., upgrading of existing wastewater treatment plants). Recently, rapidly decreas- ing membrane costs is another important driving force for the widespread application of MBRs [1]. However, membrane foul- ing and high energy consumption remain the main drawbacks. It is generally accepted that biology, membrane characteristics, configuration, and operational conditions of membrane modules all play important roles in membrane fouling control. The composition of activated sludge in MBRs is very com- plex, and includes natural organic matter (hundreds to thousands 0376-7388/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2007.03.033