Loading...

COD and ammonia removal from landfill leachate by UV/PMS/Fe2+ process: ANN/RSM modeling and optimization

Masouleh, S.Y ; Sharif University of Technology | 2022

181 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.psep.2022.01.031
  3. Publisher: Institution of Chemical Engineers , 2022
  4. Abstract:
  5. Landfill leachate is a highly contaminated liquid generated in municipal solid waste landfills. The application of sulfate radical-based advanced oxidation processes (SR-AOP) in landfill leachate treatments is emerging due to their ability to degrade both organic refractory matters and ammonia nitrogen. In this paper, application of peroxymonosulfate (PMS), activated by Fe2+ and UV was used as an economical and environmentally friendly approach for treatment of landfill leachate. Chemical oxygen demand (COD) and ammonia removals were measured as the two primary responses of landfill leachate to UV/PMS/Fe2+ treatment system. The main parameters (pH, PMS/Fe2+ mass ratio, Fe2+ dosage) affecting this system were modeled by two approaches; Response Surface Method (RSM) and Artificial Neural Network (ANN). Although RSM has an acceptable prediction performance (R2pred = 0.87–0.92), and the models are well fitted (R2 = 0.95–0.96), ANN can deliver a more precise estimate of the experimental targets (53% and 79% less root mean square error for COD and ammonia removals, respectively). The modeling results confirmed that ANN could be trained satisfactorily using data obtained from the CCD experimental design. Sensitivity analysis emphasized the importance of pH and PMS/Fe2+ mass ratio in COD removal and the significant influence of pH on ammonia removal. Multi-objective optimization and experimental results at optimal conditions confirmed that maximum COD and ammonia removal with minimum catalyst consumption are 80.8% and 25.6%, respectively. The results show that hybrid activation of PMS can remarkably enhance the removal of refractory organic matters in landfill leachate compared to the previously studied SR-AOP systems. FTIR spectra results indicate that after 60 min of treatment by UV/PMS/Fe2+ process, the carbonyl groups disappeared, and the amount of C-O-C in aliphatic ethers increased due to chemical oxidation. In addition, the transmittance of the bands corresponding to aromatic C[dbnd]C and asymmetric stretch of COO- decreased significantly, suggesting that the degree of leachate humification has changed. The BOD/COD ratio of the final effluent improved from 0.52 to 0.99, meaning that the treated leachate has higher capability to be treated via biological methods. © 2022 The Institution of Chemical Engineers
  6. Keywords:
  7. Artificial neural networks ; Peroxymonosulfate (PMS) ; Response surface methodology ; SR-AOP ; Ammonia ; Biodegradation ; Chemical oxygen demand ; Design of experiments ; Effluents ; Fourier transform infrared spectroscopy ; Iron compounds ; Leachate treatment ; Mean square error ; Multiobjective optimization ; Municipal solid waste ; pH ; Sensitivity analysis ; Sulfur compounds ; Surface properties ; Advanced Oxidation Processes ; Ammonia removal ; Chemical oxygen demand removals ; Landfill leachates ; Modeling ; Peroxymonosulfate ; Peroxymonosulphate ; Response-surface methodology ; Sulfate radicals ; Sulphate radical-based advanced oxidation process ; Neural networks
  8. Source: Process Safety and Environmental Protection ; Volume 159 , 2022 , Pages 716-726 ; 09575820 (ISSN)
  9. URL: https://www.sciencedirect.com/science/article/abs/pii/S0957582022000386