Increasing consumption rates of plastics, combined with the waste generated from their production, leads to several environmental problems. Presently, plastic recycling takes account of only about 10% of the plastic waste, which is achieved mainly through mechanical recycling. Chemical recycling methods, such as pyrolysis, could significantly increase overall recycling rates and reduce the need for the production of fossil-based chemicals. Produced pyrolysis oil can be used for the production of benzene, toluene and xylene (BTX) through catalytic upgrading or for the production of alkanes if used directly. Separation of high-value components in pyrolysis oil derived from plastic waste through traditional separation methods can be energy intensive. Organic solvent nanofiltration has been recognised as an alternative with very low energy consumption, as separation is not based on a phase transition. This work focuses on the screening of several (semi-) commercially available membranes using a simplified model mixture of pyrolysis oil obtained from plastics. Based on membrane performance, a selection of membranes was used to treat a feedstock obtained from the direct pyrolysis of plastics. This work shows that currently, commercial OSN membranes have promising separation performance on model mixtures while showing insufficient and non-selective separation at very low flux for complex mixtures derived from the pyrolysis of plastics. This indicates that OSN is indeed a promising technology but that membranes should likely be tailored to this specific application.
Keywords: circularity; organic solvent nanofiltration (OSN); plastic pyrolysis.