A random forest approach for predicting the removal of Congo red from aqueous solutions by adsorption onto tin sulfide nanoparticles loaded on activated carbon

In this work, tin sulfide nanoparticles loaded on activated carbon (SnS-NP-AC) was synthesized and characterized using various analytical techniques, such as SEM, BET, XRD, and UV–Vis spectroscopy. The impact of influential parameters such as the contact time, adsorbent dosage, pH, and initial dye concentration was investigated and optimization was carried out using random forest model. The optimized values of influential parameters i.e. pH, contact time, adsorbent dosage, and initial dye concentration were found to be 1, 4 min, 0.03 g, and 15 mg L⁻¹, respectively. At these optimized values CR achieve highest removal percentage (99%) and maximum adsorption capacity (384.6 mg g⁻¹). The experimental equilibrium data were fitted to different adsorption isotherm models i.e. Langmuir, Freundlich, Tempkin, and Dubinin–Radushkevich, among them the Langmuir model is found to be the best fitted and well suited model for evaluating and analyzing the actual behavior of adsorption process. The Kinetic experimental data were well fitted and they are in good agreement with pseudo-second-order and intraparticle diffusion model.