Optimisation of MF membrane cleaning protocol in an Indirect Potable Reuse (IPR) scheme

Abstract

The control of fouling in a microfiltration process dedicated to wastewater reuse demands an adequate chemical cleaning procedure. A study has been conducted to optimise this using Box–Behnken design (BBD) associated with multiple linear regression. Bench-scale modules, formed with hollow fibre membrane fouled through use on a pilot plant, were employed to assess the cleaning efficiency. Parameters studied were chemical type (oxidising, alkaline and acidic), reagent concentration, cleaning temperature and soak time. Oxidising reagents were found the most efficient followed by alkali and acid reagents. NaOCl at pH 10 was the most efficient cleaning reagent allowing 100% permeability recovery. An envelope of optimum parameters was defined from the bench scale study for this reagent. The model was found to accurately represent the plant performance for permeability measurements conducted at the same temperature on the bench-scale as that for plant operation. The effect and significance of the parameters were found to vary from one reagent to another one, underlining the importance of the assessing the entire parameter value range and the benefit offered by BBD for carrying out this efficiently with the fewest number of experiments.