Exploring Optimization of Zeolites as Adsorbents for Rare Earth Elements in Continuous Flow by Machine Learning...

Area of Science:

• Environmental Science
• Materials Science
• Data Science

Background:

• Rare earth elements (REEs) are critical for modern technologies.
• Efficient recovery of REEs from aqueous solutions is environmentally and economically important.
• Zeolites are promising adsorbents for REE removal.

Purpose of the Study:

• To apply unsupervised machine learning (ML) for characterizing REE adsorption by zeolites.
• To optimize adsorption-desorption cycles for enhanced REE recovery.
• To develop a predictive model for REE recovery using ML.

Main Methods:

• Unsupervised machine learning techniques, including Principal Component Analysis (PCA) and K-Means clustering.
• Continuous flow adsorption-desorption experiments.
• Development of a regression model for REE recovery estimation.

Main Results:

• ML successfully assessed adsorption results, enabling evaluation of different sorption cycle stages.
• Over 70% of REEs were removed from aqueous solutions.
• More than 80% of adsorbed REEs were recovered using an optimized desorption cycle.

Conclusions:

• Unsupervised ML provides a powerful tool for characterizing and optimizing zeolite-based REE adsorption.
• The developed regression model can predict REE recoveries.
• This study demonstrates a viable method for efficient REE removal and recovery.