Optimization of slaughter sludge-derived biochar via response surface methodology and backpropagation neural network for efficient uranium extraction from radioactive wastewater

Biochar materials are emerging as promising candidates for treating uranium-containing wastewater. In this study, the response surface methodology (RSM) and backpropagation (BP) neural network model were employed to optimize the processing parameters during the preparation of biochar via sludge aerobic fermentation products (SAPBC), thereby maximizing its potential in addressing the challenges associated with uranium-contaminated wastewater. Through systematic parameter optimization, the optimal pyrolysis conditions were determined as: pyrolysis temperature 382 °C, heating rate 5 °C/min, and holding time 2 h. Results obtained from batch adsorption experiments demonstrated that the maximum adsorption capacity of SAPBC-382 for U(VI) was 159.27 mg/g under a pH condition of 4.0. Kinetic analysis revealed the adsorption process followed pseudo-second-order kinetics ( R 2  = 0.992), while isotherm data best fitted the Langmuir model ( R 2  = 0.997), indicating monolayer chemisorption mechanism. Thermodynamic parameters confirmed the spontaneous and endothermic nature of the adsorption process. Notably, the SAPBC maintained >90 % removal efficiency through five consecutive adsorption-desorption cycles, demonstrating exceptional recyclability. Mechanism study reveals that the effective adsorption of U(VI) by SAPBC primarily stems from electrostatic attraction as well as the coordination of hydroxyl and carboxyl functional groups. Overall, this research adheres to the principle of waste control by waste not only provides an efficient solution for uranium-containing wastewater treatment but also establishes a sustainable approach for sludge valorization in environmental remediation applications.