Mo leaching rate is one of the key product properties for high molybdenum radioactive waste vitrification. Furthermore, soluble molybdenum yellow phase is easy to precipitate, resulting in the enhancement of Mo leaching. It was reported that ZnO can promote the network stability and chemical durability in borosilicate glass to some extent. Whereas, if the composition of solidification glass is complex, then whether ZnO can still improve the chemical stability is worth studying. On the other hand, due to the long period of chemical stability test, it is hoped that the Mo leaching rate can be predicted by mathematical simulation to accelerate the research process.

This study aims to investigate the effect of ZnO in borosilicate glass on Mo leaching rate in simulated high-level radioactive waste (HLW), and establish the prediction model of Mo leaching rate.

Firstly, a series multi-component borosilicate solidification glass samples with simulated high MoO₃ (mass ~3%) HLW and designed mass fractions of ZnO content of 1%, 1.5%, 3.5%, 4%, 4.5%, and 5%, respectively, were prepared for experimental test of Mo leaching rate. Then the glass structure gene modeling (GSgM) were employed to accurately simulate the glass properties using a small amount of data to establish the corresponding structural prediction model. Finally, Infrared spectrometer (IR), X-ray Diffraction (XRD), Differential Scanning Calorimetry (DSC) and thermal dilatometer (DIL) were used to test the transformation temperature, heating rate, coefficient of thermal expansion, etc. of the samples, and the model validation was also performed.

Experimental results show that Mo leaching is mainly affected by Si-O-Si rocking vibration around ~530 cm^-1 and ν_Si-O-Si vibration in Q⁴ at ~1 180 cm^-1, furthermore, B₂O₃, ZrO₂ and alkali variation influence the relative concentration of these two structural units, and then affect the Mo leaching. Mo leaching rate is enhanced with mass of ZnO over 3.5% in this complex glass system.

Model validation in this study proves that the structure model of Mo leaching rate has good simulation accuracy and statistical reliability, which are further improved after model iteration, hence can be used for the Mo-leaching prediction in the quick-screening of glass composition development.