| نویسندگان | Mohammad ali Nasseri,farzaneh esmaili,Morteza Ghadirian,Fatemeh Ghadirian,Negin Nasseh |
| نشریه | Journal of Environmental Chemical Engineering |
| شماره صفحات | 1-27 |
| شماره سریال | 14 |
| شماره مجلد | 5 |
| نوع مقاله | Full Paper |
| تاریخ انتشار | 2026 |
| نوع نشریه | چاپی |
| کشور محل چاپ | ایران |
| نمایه نشریه | ISI،JCR،Scopus |
| کلید واژه ها | Plant, mediated synthesis, Magnetic photocatalyst, Pharmaceutical wastewater, Machine learning, Heterojunction mechanism |
|---|
چکیده مقاله
The persistence of antibiotic contaminants in aquatic environments necessitates the development of efficient,
reusable, and visible-light-responsive photocatalysts for wastewater remediation. In this study, a magnetically
recoverable ternary heterostructure, BiFeO3/UiO-66-NH2/BiOI, was synthesized for the visible-light-driven
photocatalytic degradation of metronidazole (MTZ). The incorporation of UiO-66-NH2 and BiOI significantly
enhanced the physicochemical and optoelectronic properties of the heterostructure, increasing the specific
surface area from 1.12 to 11.6 m2/g and promoting broader visible-light absorption. Reduced Nyquist impedance
arc radius, photoluminescence quenching, and enhanced transient photocurrent response collectively confirmed
improved separation and migration of photogenerated charge carriers within the ternary heterojunction. Radical
trapping experiments further demonstrated the important role of reactive species and photogenerated holes in
the degradation process. Under optimized conditions, the BiFeO3/UiO-66-NH2/BiOI nanocomposite achieved
88.69% MTZ degradation, accompanied by 71.40% chemical oxygen demand (COD) and 68.30% total organic
carbon (TOC) removal, indicating substantial pollutant mineralization. The photocatalyst retained high catalytic
activity after eight consecutive cycles, demonstrating excellent magnetic recoverability and structural stability.
To complement the experimental findings, machine learning models including linear regression, Ridge, Lasso,
and artificial neural network (ANN) approaches were employed to predict MTZ removal efficiency. Among them,
the ANN model exhibited the highest prediction accuracy (R2 = 0.98), effectively capturing nonlinear relationships
between operational variables and photocatalytic performance. The enhanced activity of the ternary
heterostructure was mainly attributed to synergistic visible-light harvesting, increased surface area, efficient
interfacial charge transfer, and reactive species generation, highlighting its strong potential for antibioticcontaminated
wastewater treatment
لینک ثابت مقاله