Removal Of Heavy Metal Ions (Sb, As) From Simulated Wastewater By Iron-based Mesoporous Nanomaterials Immobilized On Graphene Oxide

Human activities and rapid industrialization bring about serious environmental problems,of which heavy metals pollution has become one of the environmental problems concerned by humans.Heavy metals are difficult to degrade in the environment and can be concentrated through the food chain to endanger human health.Antimony(Sb)belongs to heavy metals,and the toxicity of Sb(Ⅲ)is higher than that of Sb(Ⅴ).Inhalation of high contents of Sb can lead to antimony poisoning.These symptoms include vomiting,headache,dyspnea and other symptoms and death may occur in severe cases.Arsenic(As)as one of metalloid elements is often regarded as a heavy metal due to its similar toxicity to heavy metals.The toxicity of As(Ⅲ)is higher than that of As(Ⅴ).Acid mine drainage,sewage and industrial wastewater are rich in heavy metal ions.Direct discharge into the environment will seriously damage the ecological environment.Therefore,it is urgent to take several measures to treat the heavy metal ions in wastewater In this study,to reduce the concentrations of heavy metals in the environment,ordered mesoporous nanomaterials formed by graphene oxide-supported ferroferric oxide(Fe3O4/GO)and graphene oxide immobilized cobalt ferrite(CoFe2O4/GO)were used to remove Sb(Ⅲ)and As(Ⅲ)from simulated wastewater,respectively.Both the materials were successfully synthesized by chemical deposition and characterized by XPS,scanning electron microscopy(SEM),Raman spectroscopy,Fourier transform infrared spectroscopy(FTIR),superconduction quantum interference device(SQUID),N2-sorption,atomic force microscopy(AFM),transmission electron microscopy(TEM),XRD,small angle X-ray diffraction(SA-XRD).In addition,response surface methodology(RSM)combined with artificial intelligence,i.e.artificial neural network(ANN),genetic algorithm(GA),random forest(RF),particle swarm optimization(PSO),was used to model and optimize the experimental conditions,viz.temperature,initial pH,reaction time,initial concentration of removal processes.Gradient boosted regression trees(GBRT),Garson method and F test were used to analyze the importance of factors in the removal process.Finally,isothermal adsorption,kinetics and thermodynamics were used to study the adsorption characteristics of two heavy metals removal processes.RSM,ANN-PSO and ANN-GA were used to model and optimize the removal process of Sb(Ⅲ)and As(Ⅲ)in simulated wastewater The results show that,the ANN-GA is more suitable for modeling and predicting the removal process of Sb(Ⅲ)and As(Ⅲ)in aqueous solution in comparison with the ANN-PSO and RSM.F-test,GBRT and Garson method all show that contact time and temperature are the most important variables to the process of removal Sb(Ⅲ)and As(Ⅲ)from aqueous solutions,respectively.For the removal of Sb(Ⅲ)and As(Ⅲ),the adsorption processes all follows the Langmuir isothermal adsorption and pseudo-second-order kinetics,and the processes were both spontaneous and entropy-driven.XPS analysis showed that the Sb(Ⅲ)adsorption process was accompanied by redox reaction,while only the adsorption existed in process of removal As(Ⅲ).Therefore,it is feasible to remove Sb(Ⅲ)and As(Ⅲ)from aqueous solution by using Fe3O4/GO and CoFe2O4/GO with the aid of artificial intelligence technologes,respectively.