Search

This study investigates chemical cleaning mechanisms of a tubular ceramic UF membrane. The effect of cleaner type (ozone (O 3), sodium hypochlorite (NaOCl) and sodium hydroxide (NaOH)), clean in place (CIP) pH (11 vs. 12), and cleaning sequence on the removal of irreversible fouling of hydrophobic (humic acids) and hydrophilic (alginate with and without calcium (alginate + Ca+2 and alginate - Ca+2, respectively)) NOM fractions were investigated. Results showed that different NOM types responded differently to chemical cleaning. Alginate-Ca+2 and humic acids were equivalently removed by NaOCl or NaOH whereas a lower cleaning efficiency of alginate + Ca+2 was observed. Increasing the pH of NaOCl and NaOH CIP increased the removal of the chemically reversible fouling index (UMFI cr). The efficiency of NaOCl was always lower than that of NaOH at the same pH, which was attributed to surface tension (λ) differences in the CIP water and potential differences in cleaning mechanism. The ceramic UF CIP cleaning using O 3 (0.50 mg O 3 /mgC) for 1 h demonstrated higher cleaning efficiency for humic acids and alginate ± Ca+2, (%UMFI cr > 98%), than NaOCl or NaOH alone (%UMFI cr >80%). The O 3 CIP was as effective as 4 h cleaning using a sequential NaOH/NaOCl or combined NaOCl + NaOH CIP. • O 3 CIP (0.50 mgO 3 /mgC) is an effective cleaning method for ceramic membranes. • O 3 CIP is effective at removing hydrophilic and hydrophobic NOM fractions. • NaOCl and NaOH are more sensitive to water quality conditions for cleaning than O 3. • 1 h of O 3 CIP is equivalent to 4 h of combined NaOCl and NaOH cleaning. • Measuring the optimal ozonation conditions (time vs. O 3 dose) are required. [ABSTRACT FROM AUTHOR]

Highlights • Ceramic and polymeric UF fouling experiments were performed using five NOM models. • NOM fouling order was a NOM mixture ≈ BSA > sodium alginate ±Ca+2 > humic acids. • Backwash efficiency was 2× higher for the ceramic UF compared to the polymeric UF. • Hydraulic irreversibility was the dominant fouling resistance during NOM filtration. • Chemical cleaning was found to be effective for both membrane types. Abstract This research examines the effect of various NOM fractions on ceramic and polymeric UF membranes performance in terms of fouling and cleaning. Fouling experiments were performed using five model solutions, humic acid, protein as bovine serum albumin (BSA), alginate with and without calcium, and a combined NOM mixture. Two chemical agents were selected: an oxidant (NaOCl) and caustic (NaOH). Fouling and cleaning behavior were assessed using the resistance in series (RIS) model, membrane permeability, carbon mass balance, and fluorescence excitation and emission matrix (FEEM) analysis. The results demonstrated that NOM fouling order of the ceramic UF was similar to polymeric UF with the following trend: NOM mixture ≈BSA > alginate ±Ca+2 > humic acid. However, the backwash efficiency was 1.5×–2× higher for the ceramic UF in comparison to the polymeric UF, indicating a much higher hydraulic reversibility for the ceramic UF. A carbon mass balance in compliment with FEEM plots determined that NOM removal by the ceramic UF was ≈ 10% higher than the polymeric UF. Chemical cleaning was found to be effective for both membrane types. Thus, it was not possible to conclude, that the ceramic membrane demonstrated an advantage for chemical cleaning under the conditions studied. [ABSTRACT FROM AUTHOR]

Fouling research with polymeric membranes has demonstrated that various natural organic matter (NOM) fractions contribute differentially to membrane fouling behavior. However, limited studies exist analyzing the sensitivity of the MFI-UF to be used as a tool to differentiate NOM fouling components. The results here indicate that MFI-UF is a suitable tool for assessing NOM fouling. Specifically, NOM fouling potential was in the order of organic proteins (as BSA), polymers (alginate), and humic acid, respectively. Further, a mixed solution containing BSA, alginate and humic acid fouled similarly to the BSA solution indicating the high fouling potential of organic proteins in membrane systems. The MFI-UF value was found to increase by > 30% with increasing pressure (1–3 bar) and decreasing temperature (35°C–5°C). The filtered water volume was found to correlate with the MFI-UF values indicating the dependency of the method on testing conditions. Incorporating water viscosity and pressure values against normalized conditions (20°C and 2 bar) with the standard MFI-UF equation was found to be useful to estimate MFI-UF values at variable operating conditions, thus, enhances the potential application range of MFI-UF as a fouling index for NOM. [ABSTRACT FROM AUTHOR]

Abstract: The impact of support media on a submerged hollow fiber PVDF ultrafiltration membrane system filtering synthetic water with different organic and inorganic matter content was investigated using a factorial design of experiments. Support media was added into a membrane reactor where membrane fouling and water quality parameters were assessed. The addition of support media into the membrane reactor resulted in a 35–50% reduction in membrane fouling with different humic acid and kaolin clay concentrations compared to the control system with no media; however there was no apparent impact of support media in minimizing fouling by cellulose. The support media in the membrane reactor did not affect any of the water quality parameters measured; however, there was a small but measurable difference in the permeate volume and flux recovery (4% flux difference) between the two systems. This favorable difference was attributed to the presence of support media in the membrane reactor. [Copyright &y& Elsevier]