Water Journal : Water Journal April 2011
membranes & desalination water APRIL 2011 105 operation was rapidly resumed (Figure 2). Applications of the MDBR include pilot trials by colleagues for reclamation of a petrochemical plant wastewater, and a lab-scale evaluation of an anaerobic MDBR. In summary, MBRs are examples of a successful marriage of microbiology and membranes. The versatility of microorganisms provides halo-tolerant, thermophilic and anaerobic characteristics that can be beneficial in novel HRMBRs. Biofouling in MBRs Although biology can be very useful, the bad news is that membranes tend to foul, not only by inorganic scaling, but also by microbiological growth, as attested by the exponential number of publications on this topic over the past 20 years. In a typical MBR plant, there is a steady increase in trans-membrane-pressure (TMP), termed Stage 2, over some weeks, until at a certain point there is a sudden rapid increase or TMP 'jump'. In our labs, Jinsong Zhang investigated the effect of sludge retention time (SRT) on MBR fouling. At SRT of 10 days and flux 20 l/ m2hr, the jump occurred at about 200 hours of operation. With an SRT of 30 and the same flux, there was still only a slow increase in TMP after 500 hours. The MLSS was higher (10 g/L vs 7 g/L) but the number of smaller particles, the supernatant TOC and the supernatant polysaccharides were all lower. This points to their likely role in MBR membrane fouling. We have also investigated the effect of additives for fouling control. For example, polymer addition (Nalco's "membrane performance enhancer" or MPE) when dosed in reasonable quantities, extended Stage 2 dramatically, particularly at modest flux. Floc size was increased and, most importantly, there were fewer fine particles in suspension (Figure 3). When dosing was terminated, fouling increased immediately and was accompanied by an increase in extracellular polymeric substances (EPS), TOC and fines. Powdered activated carbon (PAC) as an additive (5g/L) also showed a reduced rate of TMP rise and a delayed TMP jump. The PAC appeared to have a three-fold effect. It supported up to 25% more biomass, adsorbed foulants and also physically assisted in scouring the foulant layer. Returning to our FOMBR studies, we have looked at the effect of orientation of the membrane. Placing the active skin facing the mixed liquor gave a stable but low flux. Placing the support polymer facing the mixed liquor gave an initially high flux, but it rapidly fouled as MLSS foulants penetrated the porous 'support' layer. We have now prepared a membrane with an active skin on both surfaces, and after initial flux decline it stabilises at a reasonable value. Biofouling in RO As we all know, growth of bacterial films, which are bacterial colonies held together by extracellular polymeric substances (EPS) on the active surface of the membrane, not only reduces flux and imposes an energy penalty but also necessitates more frequent downtime for effective cleaning. Figure 4 graphically shows the effect of such fouling. The image on the left is from the early days at Water Factory 21 in California, which is a water reclamation plant. Similar observations can be found for seawater RO plants. A build-up in TMP is required to maintain a constant flux (until the pump limit is reached). The TMP rise is caused by a fouling resistance and what is known as 'cake-enhanced osmotic pressure' (CEOP). The CEOP effect comes from the accumulation of salts near the membrane exacerbated by the 'unstirred' biofilm cake layer. Spiral-wound RO elements have flow channel spacers between the membranes to define the flow path of the feed liquid. The spacers improve liquid mixing and performance in clean membrane MPE MPE MPE Blank Blank Figure 4: MBR with added polymer (MPE). • Reduces fouling ~ particulary at higher ﬂux. • Increases ﬂoc size & fewer ﬁne particles. • When MPE terminated , EPS, TOC, ﬁnes and fouling increase. Figure 5: Biofouling in RO: • Signiﬁcant energy increase (kWh/m3) • Economic penalty Confocal image of bioﬁlm on RO membrane. Ridgway Bacteria and EPS in bioﬁlm – WF 21. 0.00 0.01 0.02 0.03 0.04 0.05 0.06 15 35 40 Polysaccharides Protein TOC Concentration ( mg.cm -2 ) Jv ( L.m-2 .h-1 ) No spacer Figure 3: Bioﬁlm development vs ﬂux: • Higher ﬂux causes more bioﬁlm development. • Bioﬁlm contributes both resistance and CEOP. • CEOP contributes up to 70% of TMP rise.
Water Journal March 2011
Water Journal May 2011