Chitosan-based magnetic nanobeads (CBN) were facilely synthesised and applied for selective recovery of boron from highly-saline wastewater. The CBN was characterised, and the results showed that the nanobeads are thermally and mechanically stable to withstand deterioration, had a saturation magnetisation of 65.25 emu/g, the surface area of 635 m(2)/g and nanobeads size of similar to 2.6-3.8 nm. An artificial neural network (ANN) model and Box-Behnken design (BBD) were developed and applied to study the relative effect and significance of input variables affecting boron sorption. Under optimised conditions, 89, 78 and 70% boron removal efficiencies were recorded at pH 7.0 in distilled water, highly-saline water and boron-spiked seawater respectively. The combination of intraparticle diffusion model and pseudo-second-order rate equation was applicable to describe the sorption kinetics. (C) 2017 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.