Unsupervised machine learning for sensitivity interpretation in the application of biochar derived from Haematoxylum campechianum to remove acetaminophen from aqueous solutions

abstract

The rising occurrence of pharmaceutical pollutants in water sources has prompted the need for efficient and sustainable treatment technologies. This research examines the adsorption capabilities of Activated Biochar obtained from Haematoxylum campechianum (AB-HC) in removing acetaminophen (APAP) from water solutions. A thorough characterization of the adsorbent was conducted using Scanning Electron Microscopy coupled with Energy Dispersive X-ray Spectroscopy, the Brunauer¿Emmett¿Teller method, X-ray Diffraction and Fourier Transform Infrared spectroscopy, along with determining the pH at point of zero charge (pHpzc). Batch adsorption experiments assessed the impact of initial APAP concentration, solution pH, contact time, temperature, and adsorbent dose on adsorption capacity. The Langmuir maximum adsorption capacity (Qm) of AB-HC was 84.5 mg/g at 40°C, with a pH of 6.5, an adsorbent dose of 1 g/L, and a contact time of 6 hours. The primary adsorption mechanisms included pore filling, non-electrostatic interactions such as hydrogen bonding, and ¿¿¿ interactions. Reusability tests showed a high adsorption capacity across three consecutive cycles, confirming the material's stability and regeneration potential while utilizing 50% ethanol as an eluent. Furthermore, this study performed a comprehensive exploratory sensitivity interpretation utilizing Principal Component Analysis (PCA) to pinpoint and measure the main factors affecting the adsorption of acetaminophen onto activated biochar sourced from Haematoxylum campechianum. The analysis indicated that initial concentration, dose, and temperature are key factors influencing removal efficiency. Notably, Principal Component (PC) 1 displayed a strong negative correlation (-0.633) with removal percentage, underscoring its essential role in the adsorption process. © 2025