A Technical Feasibility Study of the Recovery of Used Lubricant Oil Using Ceramic Ultrafiltration Membranes

This laboratory-scale experimental study investigated the purification level of used lubricant oil (ULO) filtration using a large variety of ceramic UF membranes, allowing for treatment at high temperatures unreachable for polymeric membranes. Varying pore sizes (5 nm, 10 nm, 30 nm, and 100 nm) were included as well as a range of materials (Al2O3, TiO2, and ZrO2). Moreover, four different grafting techniques were applied to alter the surface chemistry of the native membranes from hydrophilic to more hydrophobic or oleophilic, intending to further increase UF flux and/or retention. Benchmark native 10 nm TiO2 membranes shows a stable flux of 7 to 9 kg/h·m2 at 110 °C, strong (metal) impurity removal, and unexpected high water retention. All other membranes tested show fluxes that never exceed the ones for the 10 nm benchmark membranes, elucidating that surface chemistry does not help to improve the flux. In general, membrane performance is very similar for all membranes, except for flux and water retention. Systematically, high-flux membranes show high water retention, while very-low-flux membranes preferentially pass water. The variation in flux and water retention as a function of membrane pore size (before grafting) shows that surface chemistry only plays a role when the effective pore size becomes small. The study results allow for the selection of the best membranes for initial ULO treatment.