<p>In this study, slow pyrolysis of coconut shell at different heat treatment temperatures was used to prepare biochars that were subsequently characterizing in terms of textural properties; BET surface area, pore diameter and pore volume. The biochar samples with optimal BET surface area (sample prepared at 450 °C) was selected and used to optimize the process variables in co-pyrolysis with FeCl₃, ZnCl₂ and FeCl₃/ZnCl₂ salts resulting to the production of magnetic biochar samples. Central composite design (CCD), a tool within response surface methodology (RSM), was adopted to design the experiments. Effects of heat treatment temperature, duration of heat treatment and metal loading on the BET surface area of the magnetic biochar samples were evaluated. The optimum BET surface area obtained was 21064.421 m².g⁻¹ for Fe³⁺ - based biochar at heat treatment temperature of 327.88&#xa0;°C, duration of heat treatment of 1.627&#xa0;h and Fe³⁺ loading of 3.01 wt %, while for Zn²⁺ - based biochar, the corresponding values were 1709.55m² g⁻¹, 318.94&#xa0;°C, 1.613&#xa0;h and 2.90 wt % respectively; and for Fe³⁺/Zn²⁺ - based biochar, the corresponding values were 2130.37m².g⁻¹, 336.87&#xa0;°C, 1.885&#xa0;h and 46.733 wt % Fe³⁺ respectively. The most significant process variables were found to be metal loading for Fe³⁺ and Zn²⁺ - based magnetic biochars while the duration of heat treatment was the most significant process variable for the Fe³⁺/Zn²⁺ - based magnetic biochar. The validity of the model used was confirmed from the near convergence of the measured BET surface areas of the magnetic biochar prepared using the optimum process condition with the predicted values by the model.</p>