Cake/Biofilm Enhanced Concentration Polarization FIXEDCASE only OPTICAL Check

Discussion on calculation of maximum water recovery in nanofiltration system

Membranes are widely applied in water and waste water treatment as they provide an absolute barrier against the contaminants. Membranes are offered in wide pore size range and they are applied vastly due to their versatile and cost effective operation in dealing with wide range of streams. However, membranes, like any other filtration systems, suffer from fouling. Fouling layer, accumulation of rejected materials over time on membrane surface, is often called the main bottleneck of membrane processes. Fouling formation reduces water flux, increases energy consumption and leads to the early membrane replacement. To better control and mitigate fouling layer formation, better understanding of fouling mechanism and properties are required. Fouling properties can be categorized into hydraulic properties, mechanical properties, structural properties, chemical properties. These properties can be impacted by operational conditions, feed water quality and membrane properties. Moreover, these properties influence membrane performance parameters such as water flux, energy consumption and eventually the plant expenses. Therefore, the fouling properties, their inter-relations, their impacts on performance parameters should be further studied. We used novel modelling techniques and experimental measurements in laboratory and full-scale plants to study fouling properties and their impacts on membrane performance parameters. We also discussed the opportunities and challenges for future fouling study.Chapter 2. To evaluate the relation between structural, hydraulic and mechanical properties of fouling layer in the membrane systems, a novel method to extract these properties was developed to extracted fouling properties in a non-destructive and in-situ technique. The performance parameters of a dead-end UF system with integrated OCT imaging (in-situ) was coupled with a fully-coupled fluid-structural interaction (FSI) model. The dead-end UF was operated under a compression-relaxation cycle to evaluate how fouling properties changes under different applied pressure. Several mechanical models were evaluated to find the most suitable mechanical model to explain the fouling layer behaviour under compression-relaxation cycle in the dead-end UF. The results indicate that the hydraulic resistance of homogeneous biofilms under UF was much more affected by change in permeability than by the fouling layer thickness. Interestingly, we also found that even a poroelastic model (relatively simple model) can fairly good explain behaviour of the fouling layer in this study under different applied pressures. Compression of the fouling layer in UF systems can significantly increase hydraulic resistance of the membrane systems. In Chapter 2, a new technique was developed to extract fouling properties of the smooth surface biofilms. In Chapter 3 the new technique was further expanded to extract the mechanical properties of rough surface fouling layer under dead-end UF. We observed for the fouling layer which is fed with real surface water (i.e., river water), a dual-layer fouling structure with a thin and dense base layer and a thick and porous top layer could best explain the observed results. We also introduced a new fouling structure indicator, the fraction of exposed base layer, as a good indicator in the determination of water flux in UF systems.In Chapter 4 the chemical properties of fouling layer (e.g., composition) and their impacts on chemical cleaning efficiency in Reverse Osmosis (RO) systems were evaluated. Chemical cleaning protocols (often referred as CIP protocols) are usually developed under laboratory conditions (synthetic feed water, short-term experiments) and then are applied in the full-scale RO installations. This often leads to significant differences in CIP efficiency in the lab and full-scale installations. Thus, we compared the fouling layer properties and CIP efficiency of typical laboratory conditions RO and several full-scale RO plants. The results show that CIP efficiency in the full-scale RO plants are much lower than lab conditions RO. Later, we correlated such differences in CIP efficiency to their significantly different extracellular polymeric substance (EPS) properties. The EPS extracted from lab RO had different composition and adherence properties than the EPS extracted from full-scale RO. Therefore, we concluded that CIP protocols should not be developed under lab conditions. In the Chapter 5 we suggested a new method to better develop CIP protocols and study fouling properties with more industrial applications. We installed several new RO modules in the full-scale installation and they were operated for 30 days under identical conditions as the full-scale installation. Later, the fouling properties and CIP efficiency (in-situ measurement of permeability and pressure drop recovery) were compared between new RO modules (after 30 days of operation) and old RO modules (>2 years of operations). The new proposed fouling simulation method show promising results in both CIP efficiency results and fouling properties. Although fouling is inevitable part of filtration processes, its economic impacts on membrane systems is not well evaluated. In Chapter 6 cost of fouling in several RO and NF systems in The Netherlands has been calculated using plant performance data. All the cost factors contributing to cost of fouling such as CIP cost, energy cost, down cost were considered. We observed that for the RO plants, around a quarter of OPEX is caused by fouling, as oppose of around 10% for anoxic NF plants. The most important factor in the cost of fouling was considered the early membrane replacement cost, followed closely by additional energy cost. CIP costs have a minor contribution to the overall cost of fouling. Reuse of municipal wastewater effluent is part of solution to deal with water scarcity challenges. In Chapter 7 a fit-for-purpose approach to water reuse was proposed. We developed full techno-economic analysis on membrane-based Water reuse plant for municipal wastewater treatment effluent in the Netherlands. The impact of fouling and its properties and fouling cost have been integrated in all the membrane systems. A novel approach on design of water reuse plant has been offered to not only inherently reduce fouling impact but also increase plant robustness and water recovery. In chapter 8 a summarized and generalized conclusions of the previous chapters is presented. We also presents our suggestions and opportunities for the future membrane and fouling research.

Enhanced concentration polarization by unstirred fouling layers in reverse osmosis: Detection by sodium chloride tracer response technique

Comparison of calcium scaling in direct contact membrane distillation (DCMD) and nanofiltration (NF)

A model of estimating scaling potential in reverse osmosis and nanofiltration systems

Hybrid membrane system for desalination and wastewater treatment : Integrating forward osmosis and low pressure reverse osmosis

Colloidal fouling on a brackish water reverse osmosis (BWRO) membrane was simulated by a lab-scale plate-and-frame module in the presence and absence of air micro-nano bubbles (AMNBs). Synthetic feed water samples with the same physical and chemical properties, but various concentrations of colloidal silica (50, 100, 200, and 300 mg/L), were used for experiments. The results illustrate that colloidal fouling caused a severe decrease in permeate flux (24%–56%) and salt rejection (1.25% to 4.18%) in the absence of AMNBs. This reduction in membrane performance was attributed to the high hydraulic resistance and the cake-enhanced osmotic pressure (CEOP) of a fouled gel layer that were intensified with increase in the colloidal concentration. On the other hand, presence of AMNBs decreased the deposition rate of colloidal particles significantly and increased the porosity of fouling layer. Thus, an improvement in the membrane permeate flux (21%–40%) and salt rejection (1.2%–2.6%) were seen in the different concentrations of colloidal particles. The SEM images also confirmed the formation of loose fouling layers which were easily removed from the membrane surface by a clean-in-place (CIP) process. This research introduces AMNBs technology as an effective in-line method to control the adverse effects of colloidal fouling in reverse osmosis (RO) systems.

The reverse osmosis (RO) system is widely applied to produce reclaimed water for high-standard industrial use. Chlorine disinfection is the main biofouling control method in the RO systems for wastewater reclamation. However, researchers reported the adverse effects of chlorine disinfection which aggravated biofouling in laboratory-scale RO systems. In this study, four parallel 4-inch spiral wound RO membranes were used to study the effect of chlorine on biofouling in a pilot-scale RO system. The free chlorine dosages in four experimental groups were 0, 1, 2 and 5 mg/L, respectively. After continuous chlorination and dechlorination, the feed water entered the RO system. It was found that chlorine pretreatment caused a 1.9–36.7% increase in relative feed water pressure of the RO system, suggesting that chlorine aggravated the membrane fouling in the pilot-scale RO system. The microbial community structures of living bacteria in the feed water of the RO system were determined by the PMA (propidium monoazide)-PCR method and showed that the relative abundance of chlorine-resistant bacteria (CRB) was significantly increased after disinfection. Nine major genera which maintained higher relative abundance in experimental groups with high chlorine dosage were considered as possible key species causing membrane fouling, including Pedobacter, Clostridium and Bradyrhizobium.

Chapter 10 - Mass Transport in the Presence of a Fouling Layer

The optimal allocating pumping rate of a multi-well system for a brackish water desalination plant

Assessment of fieldscale zero liquid discharge treatment systems for recovery of water and salt from textile effluents

Long-term performance and fouling analysis of full-scale direct nanofiltration (NF) installations treating anoxic groundwater

A critical review of transport through osmotic membranes

A modified Assimilable Organic Carbon (AOC) procedure was adopted in conjunction with Heterotrophic Plate Count (HPC) method to assess the effect of Single Effect Distillation (SED) and Reverse Osmosis (RO) lab-scale systems on the biological stability of industrial water. Industrial water was collected from a local Industrial Water Works, pre-treated with alum coagulation and cartridge filtration, before being subjected to advanced water treatment. The results obtained in this study indicated that AOCs in the SED product water were in the range of 70-80 micrograms acetate-C/L, while those in the RO product water ranged from 30-40 micrograms acetate-C/L in the 15-min permeate to 55-65 micrograms acetate-C/L in the 3-hr permeate. The above findings suggested that product water of both systems were potentially biologically unstable and would likely lead to bacteria regrowth during its distribution and storage. Removal efficiencies of lab-scale RO and SED systems on AOC were as high as 90%, dependent on the concentration of AOC-NOX in the industrial water. The RO system had much higher organic removal efficiencies in terms of AOC and DOC than the SED system. Organics removed from both feed waters were found to be concentrated in the brine water and rejected water in SED and RO systems respectively.

Water is the most common substance in the world, however, 97% is seawater and only 3% is fresh water. The availability of water for human consumption is decreasing due to increasing the environmental pollution. According to the World Health Organisation (WHO), about 2.4 billion people do not have access to basic sanitation facilities, and more than one billion people do not have access to safe drinking water (Singh, 2006). Moreover, the world’s population is expected to rise to nine billion from the current six billion in the next 50 years. Chronic water pollution and growing economies are driving municipalities and companies to consider the desalination as a solution to their water supply problems. Generally, desalination processes can be categorized into two major types: 1) phasechange/thermal and 2) membrane process separation. Some of the phase-change processes include multi-stage flash, multiple effect boiling, vapour compression, freezing and solar stills. The pressure driven membrane processes, such as reverse osmosis (RO), nanofiltration (NF), ultrafiltration (UF) and microfiltration (MF), have found a wide application in water treatment (Charcosset, 2009). The energy required to run desalination plants remains a drawback. The energy limitations of traditional separation processes provided the impetus for the development and the commercialisation of membrane processes. Membrane technologies (simple, homogenous in their basic concepts, flexible in application), might contribute to the solution of most of the existing separation problems. Nowadays, membranes are used for the desalination of seawater and brackish water, potable water production, and for treating industrial effluents. RO membrane separation has been traditionally used for sweater desalination (Charcosset, 2009; Schafer et al., 2005; Singh, 2006). One of the limitations of membrane processes is severe loss of productivity due to concentration polarisation and fouling or scaling (Baker & Dudley, 1998; Schafer et al., 2005). Membrane pretreatment processes are designed to minimise the potential problems of scaling resulting from the precipitation of the slightly soluble ions. Membrane (MF or UF) pretreatment of RO desalinations plants is now a viable options for removing suspended solids, fine particles, colloids, and organic compounds (Banat & Jwaied, 2008; Singh, 2006). NF pretreatment of sweater is also being used to soften RO feed water instead of traditional softening (Schafer et al., 2005). The industrial development of new membrane processes, such as membrane distillation (MD), is now being observed (Banat & Jwaied, 2008; Gryta, 2007). In MD process feed water is heated to increase its vapour pressure, which generates the difference between the partial