Experimental and machine learning insights into the influence of aniline monomer concentration on the in situ polymerization of polyaniline on alumina supports for oil-in-water emulsion separation

• Polyaniline-coated ceramic membranes, engineered with contrasting super-wettability. • Highly efficient separation of surfactant stabilized crude oil-in-water emulsion. • Membranes showed remarkable separation effectiveness (∼98.5 %) with high flux. • An experimental and machine learning study. In this study, the concentration of the aniline monomer was varied to develop an efficient ceramic-supported polyaniline (PANI) membrane with contrasting superwettability surface features for the highly efficient separation of oil-in-water (O/W) emulsions. Aniline monomers (1, 2, and 4 wt %) were used to alter the PANI surface characteristics during in situ emulsion polymerization over an alumina support. The most homogeneous surface morphology was achieved using 4 wt % aniline monomer during in situ emulsion polymerization. The membranes were characterized using various techniques, including scanning electron microscopy, energy dispersive X-ray spectroscopy, elemental mapping, Fourier-transform infrared spectroscopy, and X-ray diffraction. The 4 wt % PANI-coated alumina membrane exhibited contact angles of ∼0° and ∼151.4° in water–air and underwater oil contact angles (final). The 4 wt % PANI-coated alumina membranes produced the highest water and crude O/W emulsion flux among the fabricated membranes (8471.7 and 4115.5 L/m² h, respectively) at 2 bar pressure. The separation efficiency of the 4 wt % PANI-coated alumina membrane was significantly higher than that of the other two membranes, reaching up to 98.2 % using an O/W emulsion at a crude oil concentration of 100 ppm. Furthermore, the long-term experimental data validated the results; a comparatively lower flux drop was observed for the 4 wt % PANI-coated alumina membrane, with a separation efficiency of ≥98 % after 5.5 h of continuous operation. Moreover, the experimental data were optimized using state-of-the-art machine learning (ML) models. The ML modelling strategy highlighted that the trained ML models, particularly the supervised vector machine model, obtained an outstanding R 2 of ∼0.96 for the separation efficiency and permeate flux models. This study demonstrates that PANI-coated alumina membranes can handle oily wastewater streams, which has important operational and environmental implications.