| Summary: | Antibiotics in nonclinical environments represent a serious risk to human health due to their role in the antimicrobial resistance. The present study aimed to optimise the detoxification of cephalexin (CFX) by the Momordica charantia extract zinc oxide nanoparticle catalyst (MCZnO
NPs) as a function of dosage of ZnO NPs, time, pH and CFX using the artificial neural network model (ANN). The effect was simulated using deep learning analysis to evaluate and explain the behaviour of CFX degradation. Interactions between these factors and the classification of the photocatalysis (low, medium, average, good and high) were analyzed using factor of principal component analysis (F, PCA), discriminant analysis (DA) and Agglomerative hierarchical clustering (AHC). MCZnO NPs have a white colour, spherical shape, non-agglomerated, smooth surface and size-wise they ranged from 50 to 100 nm. The ANN results indicated that 88.87% of CFX was degraded using 50 mg/L of MCZnO NP, 40 mg/L of CFX, at pH 9, and after 180 min. Simulation analysis revealed that MCZnO NPs were efficient in degrading CFX concentrations (up to 60 mg/L) with 100% removed depending on pH and time. The interaction between F1 and F2 was 94.59% at which pH (x2) and CFX (x4) factors exhibited a high correlation with a synergistic effect on CFX degradation, 20% of the degradation of CFX could be classified as a high percentage (> 90%). These findings reflected the role of deep learning analysis in understanding the behavior of CFX for the degradation process.
|
|---|