Functionalization of biochar using SDS/SAP nanomicelles enhanced its immobilization capacity for dyes and heavy metals in water

To enhance the adsorption capacity of biochar (BC), herein a novel multifunction modified biochar (SDMBC) was prepared by directly crosslinking of the nanomicelle of sodium dodecyl sulfate/sapindus-saponin (SDS/SAP) composite system onto the BC through a simple, environmental friendly approach. Result showed that the adsorption performance of SDMBC has been greatly improved, compared with BC or using alone SDS and SAP, adsorption ability increased by 48.83%, 29.50%, 36.44%, respectively, the best modified effect was appeared when the concentration of SAP to SDS was 0.8 and 0.8 CMC. SDMBC exhibited high adsorption abilities of 130.23, 108.43, 277.09 125.27, 112.78 mg/g for heavy metal ions lead Pb(II), Cadmium Cd(II) and organic pollutants with different chemical properties bisphenol A(BPA), Methylene blue (MB), P-nitrophenol (PNP), respectively, higher than most previously reported adsorbents, importantly, SDMBC can still efficient removal capabilities even in the binary competition. Subsequently, the SDMBC and BC was characterized by Fourier Transform infrared spectroscopy (FT-IR), X-ray diffraction (XRD), scanning electron microscopy (SEM), Zeta potential (Zeta), it found SDMBC has a more layered structure, richer functional groups and more amorphous structure compared with BC, which are closely related with improving its adsorption capacity. The adsorption behavior of SDMBC for MB show that process was found to be spontaneous, propitious, endothermic, the adsorption isotherms fitted Freundlich models well, pseudo-second-order best describes kinetics adsorption, suggesting that the process is multi-layer chemical adsorption. The little affected by ionic strength and coexisting substances, could remained removal rate over a wide pH range, SDMBC still keep high removal rates even after 5 reuses. Based on FT-IR analysis, plausible adsorption mechanism proposed, including hydrogen bond, electrostatic attraction and π-π bonding. Cost analysis manifests that the SDMBC are high efficiency and cheap eco-adsorbents compared with commercial activated carbon, and the SDMBC dosage required for the removal of 99% of a fixed amount of MB in different volumes of effluent was predicted. Seven machine learning (ML) models were used to predict the MB (60 mg/L) adsorption of the SDMBC, using Shapley Additive Explanations (SHAP) for model interpretation. Finding Extreme Gradient Boosting (XGBoost) exhibited best performance, the order of feature importance as time> Ratio> pH> concentration> temperature. Thus, SDMBC as a new cheap and eco-adsorbents, can be used to effectively remove various types of pollutants, has a great application potential in sewage treatment, while the accurate ML prediction model presented a valuable advice for designing efficient adsorbents and optimization operating conditions in the future.