Improving the Environmental Performance Of Small And Medium-Sized Manufacturing Enterprises In South Africa

Abstract: Dynamic membrane system behaviour must be adequately addressed to avoid process unfeasibility. The lack of proper analysis will mean relying on erroneous permeate flux values in the system design, which will lead to quick and/or steady high fouling rates. In this paper, the authors present additional data supporting the boundary flux theory as a helpful tool for membrane engineers to carefully avoid process failures. By fitting the dynamic permeate flux data to the boundary flux model, it was possible to calculate the β fouling index for the three selected membranes (one nanofiltration (NF) and two reverse osmosis (RO) ones). The dynamic flux given by the low-pressure RO membrane did not follow sub-boundary operating conditions, since a sharp flux loss was measured throughout the whole operating cycle, pinpointing that supra-boundary flux conditions were governing the system. This was supported by the calculated value of the β fouling parameter, which resulted to be in the order of ten times higher for this membrane. However, the values of β→0 for the SC-RO and DK-NF ones, supported by the very low value of the sub-boundary fouling parameter α (0.002 and 0.007 L·h −1·m−2 ·bar−2 , respectively), ensure nearly boundary operating conditions for these membranes.

 Introduction Industrial wastewater treatment, such as those used for tannery wastewater, is complex due to the variety of chemicals added at different stages of processing of hides and skins. Major problems in tanneries are due to wastewater containing heavy metals, toxic chemicals, chloride, lime with high dissolved and suspended salts and other pollutants (Uberoi, 2003). The tanning process and the effluents generated have already been reported in literature (Wiegant et al., 1999, Sreeram and Ramasami, 2003, Stoop, 2003). Many conventional processes were carried out to treat wastewater such as biological process (Ahn et al., 1996, Vijayaraghavan and Murthy, 1997, Wiemann et al., 1998, Di Iaconi et al., 2003, Farabegoli et al., 2004), oxidation process (Sekaran et al., 1996, Dogruel et al., 2004, Sacco et al., 2012, de Caprariis et al., 2012) and chemical process (Di Iaconi et al., 2001, Orhon et al., 1998, Song et al., 2004) etc. Among these, physical and chemical methods are considered very expensive in terms of energy and reagents consumption (Churchley, 1994, Stern et al., 2003), and generation of excessive sludge (Chu, 2001). To reduce the production of sludge by the treatment of this wastewater combined or alternative systems must be explored. In particular, in this work, two spiral wound membrane modules were used: nanofiltration (NF) and reverse osmosis (RO). The goal of this approach is to insert membranes into the cycle of wastewater treatment in order to remove the entire chain of biological treatment and the resulting post physico[1]chemical residue with a significant reduction of sludge up to 95%. A modified version of the traditional method used to measure critical fluxes of membranes, that is the pressure cycling method, was applied to measure both the critical and the threshold flux on the nanofiltration membrane in order to optimize the operating conditions. Once obtained the critical and threshold flux values, this data was used as input for a batch membrane process optimization method developed previously by Stoller at al. (Stoller and Chianese, 2006, Stoller and Bravi, 2010, Stoller, 2009, Iaquinta et al., 2009, Stoller, 2008, Stoller, 2011). The output of the method indicates the optimal permeate feed flow rate which should be used during the batch in order to inhibit membrane fouling. Finally, the obtained results were compared from an economical point of view with a conventional biological process to validate the membrane plant as possible alternative to conventional process.