Immobilization of uranium tailings by phosphoric acid-based geopolymer with optimization of machine learning

To decrease the contaminant leaching and radon exhalation from uranium tailings, a phosphoric acid-based geopolymer (PAG) precursor was selected as a solidifying agent to bind coarse sands to achieve compact structures. Machine learning was applied to explore the optimal ratio of geopolymer preparation, aimed at achieving a higher compressive strength of solidified bodies. Results showed that the maximum compressive strength of 18.964 MPa appeared at the mass ratio of 2.8 for phosphoric acid/kaolin. The uranium leaching rate of 0.70 x 10(-6) cm/d on the 42nd day was three orders of magnitude less than the clay mixture-based geopolymer solidified bodies. The successful synthesis of geopolymer was evidenced by the X-ray diffraction (XRD) and Fourier transform infrared spectroscopy (FTIR), the homogeneous and dense structure of solidified bodies was characterized by the scanning electron microscopy (SEM).

Automated summary

In this study, a phosphoric acid-based geopolymer was used as a solidifying agent to bind coarse sands and achieve compact structures, aiming to decrease contaminant leaching and radon exhalation from uranium tailings. Machine learning was employed to determine the optimal geopolymer preparation ratio, resulting in higher compressive strength of the solidified bodies.