PhD Defence Charu Chawla

Biofouling in low pressure ultrafiltration modules forĀ  point-of-use applications

Access to clean, safe drinking water is still a luxury for millions residing in developing countries. Parts of Asia and Africa are worst affected by water quality issues. Lack of water treatment infrastructure and poverty has limited the access to clean and safe drinking water in these regions that belong to the bottom of pyramid of social and economic growth. On the other hand, there are regions with water distribution networks, but due to poor maintenance and improper management, the water quality delivered by them is insufficient. Hence, to cater the needs of people without access to potable water, the World Health Organization (WHO) highlighted the potential of Point-of-Use (PoU) drinking water treatment systems operated at individual and community levels.

Charu Chawla is PhD student in the MESA+ research group Membrane Science and Technology. Her promoters are Kitty Nijmeijer and Rob Lammertink.

Ultrafiltration (UF) membranes have displayed excellent performance as PoU treatment option in termsĀ  of pathogen removal, ease of maintenance and cost effectiveness. Still, the performance of UF membranes is hampered by fouling and biofouling. In this work, we studied long term fouling development in low pressure, gravity driven (<1 bar) hollow fiber based ultrafiltration modules simulating PoU treatment systems. The key findings of the present study are highlighted below.

In general, the performance of UF can be affected by three major factors: operational conditions, feed water quality and membrane properties. Within this scope, various parameters that could influence fouling development in low pressure UF are studied. In the first part of the study presented in Chapter 2, various operational modes are investigated such as inside/out vs. outside/in operation and discontinuous vs. continuous filtration, to study their effects on fouling and permeate production with surface water as feed. Inside/out UF membranes, with the separating layer on the bore side, are tested both in inside/out as well as in outside/in mode of operation. Flux stabilization is observed for these hollow fiber UF, operated at gravity pressures of 0.1 bar. Flux values stabilize around 2 L/m2.hr in almost all cases (40 times lower than the clean water permeability values provided by the manufacturer). In most experiments the fouling layer mainly consists of diatoms, inorganic particles and a few microbial clusters. Cake layer formation is the main fouling mechanism for inside/out operation, while pore blocking dominated the outside/in operational mode. The intermittent inside/out mode of operation resulted in lower hydraulic resistances compared to outside/in due to the lower foulant load per unit membrane area as well as back diffusion of foulants into the bulk solution . It is shown that the PoU systems investigated can be operated for longer durations (>30 days) without the need for strong chemical cleaning. Overall, the intermittently operated inside/out configuration performed better than continuously operated membrane systems and the outside/in operational mode.

In the second part of the work, presented in Chapter 3, the effects of feed water quality on fouling development are studied by challenging the in-house constructed PoU with the secondary wastewater effluent (WWEf) as feed and comparing inside/out (I/O) vs. outside/in (O/I) operational modes. The findings shine light on the underlying principles regarding biofouling development in low pressure hollow fiber membranes using WWEf as the feed solution. While the flux declines monotonically for outside/in (O/I) mode, a fluctuating flux trend with fluxes decreasing and increasing over time is observed for the inside/out (I/O) operational mode. It is shown that the fluctuations in the nematode population correlate well with the periodic flux trend in the I/O mode. The biofouling layer accumulated on the membrane surface consists of biotic as well as abiotic components. It is proposed that the predator-prey relationship between nematodes (metazoas) and the microbial community present in secondary WWEf strongly influences the flux behavior. For our system, a threshold nematode population density of 2.2x106 counts/m2 was required to influence the permeate flux behavior. The population of adult nematodes has a larger influence on permeate production compared to the number of juveniles. The effect of seasonal variation on nematode population density and growth is evident. During summer season, at ambient temperatures of 25oC, a higher population density of nematodes is seen within the fouling layer, whereas at temperatures of 12oC or below, the growth of nematodes is retarded and hence population density decreases. It is shown that the presence of active nematodes can lead to a 50% - 75% increase in fluxes. Predation by biologically active and fully grown nematodes is suggested as a biofouling mitigation strategy during low pressure membrane filtration. The scientific findings based on the impact of naturally occurring biological interactions of bacterivorous nematodes on membrane biofouling can be helpful in engineering membrane technology. The natural predator-prey interactions can be utilized for designing more efficient and self-sustainable Point-of-Use systems.

Chapters 2 and 3 focused on utilizing gravity pressures for ultrafiltration, as this is relevant for the regions where water supply infrastructure is not in place. However, countries like India and China do have piped water supply, especially in the urban areas. Therefore, in the other sections biofouling development is tested on UF membranes under tap pressures. Continuing with the study on the effect of feed water quality, biofilm formation is studied under varying monovalent to divalent (M:D) cationic ratios in the feed water. The findings are discussed in Chapter 4. Biofilm formation is evaluated in terms of hydraulic resistance and production of EPS components such as polysaccharides and proteins. A natural consortium of bacteria native to drinking water is used to grow a biofilm on at sheet ultrafiltration membranes operated at constant pressures of 0.5 bar. The three different monovalent to divalent (M:D) ratios investigated are 1.31:1, monovalent (M) ions only (no divalent ions), and 2.23:1. The removal of divalent cations from feed water considerably reduces the production of extra cellular polymeric substances (EPS), and especially that of polysaccharides. The hydraulic resistances of biofilms are the lowest in the absence of divalent cations, i.e. with monovalent ions only. Though the presence of magnesium as a divalent cation at M:D 2.23:1 promotes the secretion of polysaccharides, the hydraulic resistances are highest at M:D 1.31:1 in combination with the presence of calcium. In the final part of the study presented in Chapter 5, membrane surface properties are modified by altering surface charges (positive as well as negative) via layer by layer (LbL) deposition of the polyelectrolytes poly(diallyldimethylammonium)chloride (PDADMAC) and poly(styrenesulfonic acid) (PSS) on UF membranes. The biofouling potential of modified membranes is evaluated by the growth of drinking water bacteria on the membrane surface under dynamic filtration conditions. The results show that the performance of coated membranes is hampered by biofouling with no effect of the positively or negatively charged membrane surface on biofouling potential for drinking water filtration. The bacterial deposition on both positively charged and negatively charged surfaces is confirmed by SEM and ATP analysis. However, with the inclusion of a daily cleaning step (forward flushing) to simulate realistic operating conditions, we observe much higher flux recoveries for the coated membranes compared to the uncoated ones. Though bacterial attachment is similar for both coated and uncoated membranes, the presence of highly hydrated polyelectrolyte multilayers (PEM) on coated membranes weakens the interaction between the bacteria and membrane surface, thereby preventing the irreversible adhesion of bacteria on the surface. Consequently, the biofouling layer is easily removed from these membranes by this simple physical cleaning method. The outcome of the present study clearly shows the importance of conducting bacterial adhesion tests on modified surfaces under dynamic conditions.