Elucidating morphological effects in membrane mineral fouling using real-time particle imaging and impedance spectroscopy

Mineral fouling is a major hindrance to high recovery effluent nanofiltration, with calcium phosphate (Ca-P) and calcium carbonate (CaCO3) the most prevalent mineral foulants. In this study, we used a novel combination of real-time in-line microscopy, electrical impedance spectroscopy (EIS), post SEM analysis, and filtration metrics (water flux and rejection) to study mineral fouling mechanisms of Ca-P and CaCO3 salts in synthetic effluent nanofiltration. We used nanofiltration (NF) polyelectrolyte multilayer (PEM) membranes, prepared by static layer-by-layer (LbL) coating of a cationic polymer - polydiallyl dimethylammonium chloride, and anionic polymer - poly styrenesulfonate (six bi-layer) on a polyethersulfone (PES) ultrafiltration (UF) membrane. Increasing permeate recovery over filtration time was simulated through additions of CaCl2 with NaHCO3 or NaH2PO4/Na2HPO4. Using the novel combination of methods, we delineated the mechanisms governing fouling development with time for both CaCO3 and Ca-P. For CaCO3, a transition from heterogeneous precipitation on the membrane surface (scaling) to particulate fouling due to bulk precipitation was identified. For Ca-P, a transition from fouling by amorphous particles to fouling by crystalline particles was identified; and this phase-change was captured in real-time images using an in-line microscope. We also found that for similar precipitation potentials measured by weight, Ca-P fouling was more detrimental to water flux (86% decrease) compared to CaCO3 (20% decrease) due to the voluminous amorphous phase. We established in-line microscopy as a new useful method to study mineral fouling, as it gives invaluable information on the suspended particles in real-time. Combining it with EIS gives complementary information on mineral accumulation on the membrane surface. Insight from this study and further use of these methods can guide future strategies towards higher effluent recovery by membrane filtration.

Concentration of Municipal MBBR Effluent by FO for Resource Recovery: Batch Experiments in Side-Stream Configuration

A novel approach for resource recovery includes forward osmosis (FO) as a concentration step in municipal wastewater treatment. The current study investigates different pre-treatment strategies including biological treatment with a moving-bed bioreactor (MBBR) at different loading rates and particle removal by filtration and sedimentation. Membrane performance and recovery potential for energy and nutrients were investigated in laboratory-scale FO experiments in batch mode using pre-treated municipal wastewater as feed and 35 g/L NaCl as a draw solution. Initial water fluxes were in the range of 6.3 to 8.0 L/(m2·h). The baseline fluxes were modelled to account for flux decline due to concentration effects and to enable the prediction of flux decline due to membrane fouling. Fouling-related flux decline varied from 0 to 31%. Both organic fouling and precipitation of CaCO3 and CaHPO4 were identified by using SEM–EDS. High-rate flushing resulted in complete flux recovery under most conditions. Scaling could be avoided by lowering the pH. Two operation strategies were tested to achieve this: (1) applying a bioreactor with a low organic loading rate to achieve high nitrification, and (2) adding a strong acid. A low organic loading rate and the use of additional particle removal were efficient measures that reduced organic/particulate fouling. The recovery potentials for COD and phosphorous in FO concentrate were close to 100%.

Development of a quantification method for fouling deposits using phosphorescence

Particulate fouling on structured surfaces is typically quantified using the integral thermal or mass-based fouling resistance. The observed geometries may be structures that can improve the heat transfer in heat exchangers (e.g., dimples), cavities in components, or more complex geometries. However, due to limited accessibility or the requirement for a locally resolved measurement, the existing quantification methods may not be applicable to structured surfaces. For this reason, a new method is needed for the quantification for fouling deposits. In this study, dimpled surfaces were evaluated by measuring the integral thermal and mass-based fouling resistance and comparing it with the local fouling resistance inside and around the dimple. This comparison was carried out online with the Phosphorescent Fouling Quantification method developed for this purpose, using phosphorescent particles to quantify the deposited mass. The mass-based fouling resistance can be calculated using computer-aided image analysis. The measurements for the evaluation were conducted on dimpled surfaces, which produced a characteristic fouling pattern. With the new method a reduced surface coverage from up to 33.3 % was observed, which led to lower fouling resistances downstream of the dimple compared to a plain surface. These results confirm earlier numerical and experimental findings, suggesting an advantage of dimpled surfaces over other surface structures with respect to thermo-hydraulic efficiency as well as reduced fouling. Thus, the Phosphorescent Fouling Quantification method provides the possibility of calculating values for local fouling resistances on structured surfaces, as well as the possibility of optimizing surface structures to minimize fouling propensity.

Assessing Pretreatment Effectiveness for Particulate, Organic and Biological Fouling in a Full-Scale SWRO Desalination Plant

In this study, the removal of particulate, organic and biological fouling potential was investigated in the two-stage dual media filtration (DMF) pretreatment of a full-scale seawater reverse osmosis (SWRO) desalination plant. Moreover, the removal of fouling potential in two-stage DMF (DMF pretreatment) was compared with the removal in two-stage DMF installed after dissolved air floatation (DAF) (DAF-DMF pretreatment). For this purpose, the silt density index (SDI), modified fouling index (MFI), bacterial growth potential (BGP), organic fractions and microbial adenosine triphosphate (ATP) were monitored in the pretreatment processes of two full-scale SWRO plants. Particulate fouling potential was well controlled through the two stages of DMF with significant removal of SDI15 (>80%), MFI0.45 (94%) and microbial ATP (>95%). However, lower removal of biological/organic fouling potential (24–41%) was observed due to frequent chlorination (weekly) of the pretreatment, resulting in low biological activity in the DMFs. Therefore, neutralizing chlorine before media filtration is advised, rather than after, as is the current practice in many full-scale SWRO plants. Comparing overall removal in the DAF-DMF pretreatment to that of the DMF pretreatment showed that DAF improved the removal of biological/organic fouling potential, in which the removal of BGP and biopolymers increased by 40% and 16%, respectively. Overall, monitoring ATP and BGP during the pretreatment processes, particularly in DMF, would be beneficial to enhance biological degradation and lower biofouling potential in SWRO feed water.

Monitoring Biofouling Potential Using ATP-Based Bacterial Growth Potential in SWRO Pre-Treatment of a Full-Scale Plant

Several potential growth methods have been developed to monitor biological/organic fouling potential in seawater reverse osmosis (SWRO), but to date the correlation between these methods and biofouling of SWRO has not been demonstrated. In this research, the relation between a new adenosine triphosphate (ATP)-based bacterial growth potential (BGP) test of SWRO feed water and SWRO membrane performance is investigated. For this purpose, the pre-treatment of a full-scale SWRO plant including dissolved air flotation (DAF) and two stage dual media filtration (DMF) was monitored for 5 months using BGP, orthophosphate, organic fractions by liquid chromatography coupled with organic carbon detection (LC-OCD), silt density index (SDI), and modified fouling index (MFI). Results showed that particulate fouling potential was well controlled through the SWRO pre-treatment as the measured SDI and MFI in the SWRO feed water were below the recommended values. DAF in combination with coagulation (1–5 mg-Fe3+/L) consistently achieved 70% removal of orthophosphate, 50% removal of BGP, 25% removal of biopolymers, and 10% removal of humic substances. Higher BGP (100–950 µg-C/L) in the SWRO feed water corresponded to a higher normalized pressure drop in the SWRO, suggesting the applicability of using BGP as a biofouling indicator in SWRO systems. However, to validate this conclusion, more SWRO plants with different pre-treatment systems need to be monitored for longer periods of time.