Optimizing Methylene Blue Removal Using Magnetic Chitosan Carboxymethyl Cellulose Multiwalled Carbon Nanotube Composite

Strategic Insights -

Researchers from Yasuj University of Medical Sciences in Iran have made a breakthrough in developing an efficient method for removing methylene blue from wastewater using a novel magnetic chitosan-carboxymethyl cellulose/multiwalled carbon nanotubes composite. The study employed genetic algorithms and regression techniques, including Gradient Boosting Regressor and Maximum Likelihood Estimation, to optimize the removal process.

Key Takeaways:

  • The Gradient Boosting Regressor model showed a slightly better accuracy of 0.99, Root Mean Square Error of 0.68, and Mean Absolute Error of 0.49 compared to the Maximum Likelihood Estimation model, indicating its superiority in predicting methylene blue removal efficiency.
  • The Genetic Algorithm analysis revealed that the model has converged to the optimal solution effectively, with the best solution for X1 = 49.41, X2 = 110.62, X3 = 11.85, and X4 = 20, which gives the maximum removal efficiency of 94.64% of methylene blue.
  • The feature importance analysis showed that X2 (initial methylene blue concentration) has the highest importance, while X1 (contact time) and X4 (adsorbent amount) are less important and can be eliminated from the models.
  • The study demonstrated the stability of the adsorbent in residuals, with a Mean Residual of 0 and Root Mean Square Error of 0.68 in the training set, and 0.15 and 2.33, respectively, in the testing set.
  • The results indicate that the Gradient Boosting Regressor algorithm is more efficient and has a higher accuracy margin than Maximum Likelihood Estimation, making it a preferred choice for modeling methylene blue removal.
  • The researchers concluded that the adsorbent showed a higher removal efficiency of 94.64%, making it a potential candidate for removing dyes from wastewater.

Statistics:

  • The accuracy of the Gradient Boosting Regressor model is 0.99, while that of Maximum Likelihood Estimation is 0.94 in training and 0.95 in testing.
  • The Root Mean Square Error (RMSE) of the Gradient Boosting Regressor model is 0.68 in the training set and 0.15 in the testing set.
  • The Mean Absolute Error (MAE) of the Gradient Boosting Regressor model is 0.49 in the training set and 0.68 in the testing set.
  • The maximum removal efficiency achieved by the adsorbent is 94.64%.
  • The feature importance analysis shows that X2 has a positive coefficient of 0.72 for improved removal efficiency, while X3 has a positive coefficient of 0.66.

Sources:

  • Scientific Reports, 2025;15(1):20705.
  • Nature Portfolio, Heidelberger Platz 3, Berlin, 14197, Germany.
  • Yasuj University of Medical Sciences, Yasuj, Iran.
  • Saeid Fallahizadeh, Dept. of Environmental Health Engineering, Faculty of Public Health, Yasuj University of Medical Sciences, Yasuj, Iran.
  • Mahmood Yousefi, Yosra Maleki, Amir Sheikhmohammadi, and Alieh Rezagholizade-Shirvan.