Polyphenol adsorption and recovery from olive mill wastewater: a model reduction-based optimization and economic assessment

Adsorption is a promising technology for the removal and recovery of high-value products from waste streams. However, only a limited number of studies have focused on optimizing this process in terms of design variables and operational conditions. In this study, we performed tests of Phenolic Compounds (PC) recovery from Olive Mill Wastewater (OMW) by adsorption, followed by a model-based economic optimization of the process, using surrogates to drastically reduce the computational cost of the procedure. The Polynomial Chaos Expansion (PCE) model reduction technique was employed to obtain an efficient description of the process of PC adsorption from OMW based on a limited set of high-fidelity simulations developed in COMSOL Multiphysics. In particular, we derived PCEs to predict key performance metrics – resin utilization efficiency and adsorption yield – in a range of variability of the design variables, thus obtaining a surrogate model-based optimization of the process in terms of column height (9 m), OMW superficial velocity (25 m/h), and PC breakpoint (10%). These optimized conditions led to a high-performing PC recovery process (adsorption yield 98%, resin utilization efficiency 51%) that resulted attractive for investors, with an 8% Internal Rate of Return and a 152,049 € Net Present Value. A sensitivity analysis indicated that the process proves economically attractive at acceptable PC selling prices (3-10 €/kg PC) for PC concentrations in the treated OMW > 1 g/L. This study confirms the high potential of adsorption for recovering added-value products from wastewater in a circular economy perspective.