Early warning of biofouling in spiral wound nanofiltration and reverse osmosis membranes J.S. Vrouwenvelder a,b, ⁎, M.C.M. van Loosdrecht b , J.C. Kruithof a a Wetsus, Centre of Excellence for Sustainable Water Technology, Agora 1, P.O. Box 1113, 8900 CC Leeuwarden, The Netherlands b Delft University of Technology, Department of Biotechnology, Environmental Biotechnology Group, Julianalaan 67, 2628 BC Delft, The Netherlands abstract article info Article history: Received 21 May 2010 Received in revised form 19 July 2010 Accepted 20 July 2010 Available online 24 August 2010 Keywords: Biofouling monitor Biofilm Feed spacer channel Linear flow velocity NF RO Costs In spiral wound nanofiltration and reverse osmosis installations several fouling types may occur. Simultaneous screening of all fouling types could be carried out to establish the impact of each individual fouling type on membrane performance. In extensively pre-treated water biofouling is the major fouling type. Membrane manufacturers recommend to take corrective actions based on a 15% pressure drop increase criterion. In general this approach is not successful. For an adequate anti-biofouling strategy early warning monitoring plays an essential role. Early warning of biofouling requires (i) a Membrane Fouling Simulator (MFS) supplied with feed water of the membrane filtration installation, (ii) a sensitive differential pressure drop transmitter over the MFS to monitor the pressure drop increase, and (iii) a higher linear water velocity in the MFS compared to practical conditions to increase the biofilm formation rate and pressure drop, enabling earlier detection. Action based on this early warning monitoring system for biofouling control is more reliable and successful than the approach recommended by membrane manufacturers and the costs are a fraction only of the potential annual savings. © 2010 Elsevier B.V. All rights reserved. 1. Introduction An early warning monitoring system should be an essential part of an adequate anti-biofouling strategy for reverse osmosis (RO) and nanofiltration (NF). Earlier detection of biofouling enables corrective actions at an earlier stage. This strategy is expected to be more effective than control actions after a pressure drop increase of 15% over the total installation, as recommended by membrane manufac- turers, when already much biomass is accumulated at the feed side of the membrane installation. The feed water quality plays a major role in membrane biofouling. This suggests the use of a biological parameter to assess the biofouling potential of the feed water. Vrouwenvelder et al. [1] evaluated the use of feed water parameters such as adenosinetriphosphate (ATP) [2] and total direct cell counts [3], representing the concentration of microorganisms, Assimilable Organic Carbon (AOC) [4] as a measure for growth promoting substances, and the Biofilm Formation Rate (BFR) in a glass-ring biofilm monitor [5]. These water quality parameters ATP, total direct cell counts, AOC and the BFR were not sensitive enough to be used for early warning of biofouling [1]. In extensively pre-treated water the pressure drop increase over the membrane modules in the installation is directly related to membrane biofouling [6,7]. Standard pressure drop measurements commonly used in practice are not sensitive enough to quantify biofouling timely and therefore cannot be used as an early warning system enabling control actions [1]. More sensitive and accurate differential pressure drop transmitters – than commonly used standard pressure equipment – are required for early warning of biofouling. In addition to sensitive differential pressure transmitter use early warning monitoring of biofouling also requires conditions facilitating more early detection such as a higher linear water velocity causing a higher biofouling accumulation rate. Recently, interest in early warning system for biofouling has been growing [8,9]. Several methods for monitoring of biofouling/biofilm accumulation have been described [10–21]. The ideal tool for biofouling monitoring in spiral wound membrane systems has to meet a large number of requirements [12,13,16], e.g. representative- ness and sensitive in-situ non destructive assessment of performance loss and fouling. However, no available monitor was fulfilling the combination of requirements. Therefore, a tool was developed for the validation of membrane fouling: the Membrane Fouling Simulator [22–24]. The major advantages of the MFS are representativeness for spiral wound membranes and the small size requiring small amounts of water and chemicals. Using the MFS, fouling can be monitored by (1) operational parameters like pressure drop, (2) non-destructive (visual, micro- scopic) observations using the sight glass and (3) analysis of coupons sampled from the membrane and spacer sheet in the MFS. Desalination 265 (2011) 206–212 ⁎ Corresponding author. Wetsus, centre of excellence for sustainable water technology, Agora 1, P.O. box 1113, 8900 CC Leeuwarden, The Netherlands. E-mail address: Hans.Vrouwenvelder@wetsus.nl (J.S. Vrouwenvelder). 0011-9164/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.desal.2010.07.053 Contents lists available at ScienceDirect Desalination journal homepage: www.elsevier.com/locate/desal