Magnetic separation of micron-sized particles: process study and regression modelling using moving least squares and multivariable power least squares method

Abstract

The presence of fine particles has reduced the quality of water resources. Unfortunately, the fine particles normally take a longer time to be removed. Realizing this, the present study aims to employ a magnetic adsorbent to assist a fast separation of the fine particles from their solution. Here, micron-sized SiO₂ particles were used as the model fine particles; meanwhile, the magnetic adsorbent was designed by functionalizing iron oxide nanoparticles (IONPs) with chitosan. A process study was done based on the effects of medium pH, IONPs concentration, and contact time. Results implied that both bare IONPs and chitosan functionalized-IONPs (denoted as CF-IONPs) were best in removing SiO₂ at pH 4 owing to the electrostatic attraction force. However, as compared to the bare counterpart, CF-IONPs is more suitable for large scale application as it is effective in a wider pH range, requires lesser adsorbent dosage and shorter adsorption time. Despite that, removal of SiO₂ using both types of adsorbents were found to follow pseudo-second order kinetic. The obtained data was modeled using either the moving least squares (MLS) or multivariable power least squares (MPLS) method. In particular, the indices of each parameter obtained from the MPLS-generated equation indicated that the most dominant parameter that governing the SiO₂ removal by bare IONPs is medium pH, while the one for CF-IONPs is contact time.

Keywords:

• Fine particles
• magnetic nanoparticles
• moving least squares method
• multivariable power least squares method
• separation
• water treatment

Disclosure statement

All authors declare that there are no competing interests associated with this manuscript.

Additional information

Funding

Open Access funding provided by the Qatar National Library. The authors express the gratitude to UCSI University for financial support through Research Excellence & Innovation Grant (REIG) with project code REIG-FETBE-2020/015.