Compositional data analysis – CoDA, as a tool for evaluating arsenic adsorption on sludge from a thermally modified drinking water treatment facility

Arsenic (As) contamination in water is a public health problem worldwide, impacting countries like Peru where recorded concentrations exceed permissible limits. Biosorption by the metal oxide-containing sludge produced by Drinking Water Treatment Plants (DWTPs) may be a practical way to reduce arsenic pollution in water. This study evaluated the adsorption of As on thermally modified sludge from a DWTP (Arequipa, Peru), and incorporated Compositional Data Analysis (CoDA) to ensure a robust statistical treatment of removal efficiency. Physicochemical and structural characterization revealed a mesoporous nature and high contents of Fe and Al oxides. Thermal activation at 300 °C (SA300) generated the highest specific surface area ( S BET = 66.03 m 2 /g) and removal efficiency. Adsorption occurred mainly in the first 180 min of the process, reaching maximum yield at 507 min (R = 93.2%). The compositional model confirmed the optimal activation temperature (T = 300 °C). A compositional Box Behnken design was used for optimization of the adsorbent dose, the initial concentration of As (CoAs) and the solution pH achieving an R = 97.9% with a C 0 As = 10 mg/L optimal configuration, dose D = 13 g/L and pH ≈ 7.5. The kinetic data were fitted to the pseudo-second-order model, indicating a chemisorption mechanism. The Langmuir model showed the best fit (R 2 = 0.998, K L = 0.3 L/mg, q max = 9.862 mg/g) over Freundlich’s, confirming a monolayer adsorption. Overall, CoDA showed its usefulness in adsorption experiments and offered methodological advantages over conventional techniques.