Efficient adsorption of methylene blue (MB) by an eco-friendly chitosan derivative adsorbent: An RSM and ANN modeling study
This paper investigates the use of response surface methodology (RSM)/central composite design (CCD) and artificial neural networks (ANN) to develop a method for removing MB from synthetic wastewater using a derivative of chitosan. On this note, chitosan adsorbent material was developed by cross-lin...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Desalination and Water Treatment |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1944398624204537 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper investigates the use of response surface methodology (RSM)/central composite design (CCD) and artificial neural networks (ANN) to develop a method for removing MB from synthetic wastewater using a derivative of chitosan. On this note, chitosan adsorbent material was developed by cross-linking and grating. The materials were characterized using different sets of analytical techniques, including scanning electron microscope (SEM), X-ray diffraction (XRD), Thermogravimetric analysis (TGA)/Differential thermal analysis (DTA), Fourier transform infrared (FTIR), and Brunauer-Emmett-Teller (BET)/Barrett-Joyner-Halenda (BJH). However, during the adsorption investigations, RSM-CCD was used to design the experiment. One neuron was employed as the output layer, which corresponds to the removal efficiency of MB, and variables, including adsorbent dosage, pH, contact time, and MB concentration, were taken into consideration as the input layer feed data. Statistical metrics such as average relative errors (ARE), coefficient of determination (R2), mean squared error (MSE), Pearson's Chi-square (χ2), root means square errors (RMSE), and the sum of squares of errors (SSE) were used to measure the RSM and ANN models. It was discovered that the highest removal percentage for MB adsorption from RSM result was 97.7 %, at various conditions of pH 7, concentration of 125 mg/L, adsorbent mass of 6.0 g, and contact time of 55 min. The isothermal studies revealed that the experimental data fit the Langmuir model well, with a maximum adsorption capacity of 128.67 mg/g. The ideal trained neural network represents the training, validation, and testing phases, with R2 values of 0.99964, 1, and 1, respectively. However, the results demonstrated that the ANN results outperform the RSM-CCD model approach. These results imply that ANN can be applied to predict the amount of MB removed from wastewater. Five successive adsorption/desorption studies were performed, and it was found that the beads could be regenerated. |
|---|---|
| ISSN: | 1944-3986 |