Acid mine drainage treatment by nanofiltration membranes: impact of aluminium and iron concentration on membrane performance.

Abstract

The purpose of this study is to evaluate the rejection of different elements in acid mine drainage (AMD), specifically metals, by means of nanofiltration (NF) membranes. AMD is characterised by its high acidity and high concentration of metals and sulphate. These aspects make AMD treatment a point of interest for industry and society. The main reasons for that are that being able to properly treat AMD and separate the different metals dissolved allows its recovery and reuse. Furthermore, proper treatment via NF increases water quality, as it decreases the concentration of dissolved metals, diminishing the effect they have on animals and plants when discharged into the environment. Experiments were carried out in a laboratory-scale plant, based on a flat sheet membrane with recirculation of permeate and concentrate to the feed tank. A total of 9 experiments were designed, in order to study rejection of different species varying pH solution and concentration of aluminium or iron. The solutions were prepared to resemble AMD, including different metals (aluminium, zinc, copper…) and rare earth elements (REE) (lanthanum, neodynium…). Sample composition was analysed via ICP by the Institut de Diagnosi Ambiental i Estudis d’Aigüa (IDAEA) of the Consejo Superior de Investigaciones Científicas (CSIC). After obtaining the rejection, data were modelled using the Solution-Diffusion-FilmModel, plotting dominant salt and trace ion rejection depending on the trans-membrane flux. This model also allowed permeability of single trace ions to be calculated, comparing experimental and theoretical values. The results obtained show that addition of metals in the solution affect rejection of the species in the solution, as sulphate forms complexes with metal ions. As iron and aluminium concentrations increased, rejection was generally increased. This also supports the theory that divalent ions show higher rejection than monovalent ions. Finally, experimental costs and environmental impact were addressed