Classical and Bayesian Computational Statistics Applied to Acetylsalicylic Acid Mitigation in Aqueous Solutions: Study of Batch Scale, Kinetic Modeling, and Fixed-Bed Column Adsorption

This study focused on the adsorption of acetylsalicylic acid (ASA) using activated carbon, alongside an approach integrating Monte Carlo Bayesian modeling. The influence of the adsorbent mass, contact time, and pH was studied in batch adsorption. A fixed-bed column study was conducted to evaluate practical conditions such as the feed-flow rate, adsorbent bed, and initial ASA concentration. The Monte Carlo Bayesian method was applied to analyze the parameter estimation procedure and to improve our understanding of the underlying mechanisms of adsorption. The efficiency of ASA adsorption achieved was 2.5 mg g–1. The kinetic model of the pseudo-first-order (PFO) presented a lower Bayesian information criteria (BIC) value (113.72). The diffusion of the column bed (Dz) and the Langmuir constant (KL) presented significant variation when the Dz varied in the range of [50.71–142.45] and the KL varied in the range of [0.020–0.059]. The results indicate that the model used is robust enough to simulate the breakthrough curves even with a low coefficient of determination.