Modelling copper adsorption on olivine process dust using a simple linear multivariable regression model

This paper investigates the adsorption of copper on process dust from the production of olivine aggregates as a function of pH and the concentration of copper in solution. The olivine process dust, containing more than 90% forsterite, was retrieved from the dust removal system at A/S Olivin’s processing plant at Åheim in western Norway. Following characterisation of the material, batch adsorption experiments were conducted at 25 °C at an ionic strength of 0.05 M using a fixed adsorbent dose of 10 g/l and initial copper concentrations (Cu]_i) ranging from 20 to 600 μM (i.e. 1.27–38.1 mg/l). The final pH in solution (pH_f) was varied from approximately 4 to 6. The obtained results show that the olivine process dust is capable of greatly reducing the concentration of copper in solution, even when exposed to relatively high initial copper concentrations. A linear multivariable regression model (LMVR model) was fitted to the logarithmically transformed experimental data. The good fit of the LMVR makes it possible to estimate retention as a function of pH_f and the final copper concentration (Cu]_f), or, alternatively, Cu]_f as a function of pH_f and Cu]_i. From an environmental engineering point of view, the latter would often be of greater interest since the success criteria usually are related to the resulting concentrations in solution.