Simulation of agglomeration/defluidization inhibition process in aluminum–sodium system by experimental and thermodynamic approaches

This study modifies a defluidization time prediction model under the influence of agglomeration inhibition provided by the addition of Al into a fluidized bed system. Operational parameters, such as the Al/Na ratio, temperature, gas velocity, and the particle size of bed materials, were considered by the prediction model developed by Lin et al., and thermodynamic consideration of the Al–Na–Si reaction system revealed the agglomeration mechanism. An exponential decay of fN was found with the increase of the Al/Na ratio, and the data analysis gave a regression formula of fN = 0.0601 + 0.4983 ∗ exp [− 4.4183 ∗ (Al/Na ratio)], with an R2 of 0.9972. The simulation data generally match well with the experimental results, particularly at lower operation temperatures and Geldart B group's particles, with acceptable average errors of − 8.4% and − 4.9%, respectively. However, relatively poor matches were found in the cases with different gas velocities, a higher operation temperature (900 °C), and a larger particle size (Geldart D group) of bed materials.

Graphical abstractThe aim of this study is to modify the defluidization time model developed by Lin et al. with experiments measured under different Al/Na ratios. The results were used in the model to obtain the ratio of Na in bed materials (fN), and the change of fN was modified in the prediction model when Al was added in the waste.The results of f-value regression by calculation of experimental results at different Al/Na ratios. (Operating conditions: temperature: 800 °C, particle size: 770 μm, gas velocity: 0.18 m/s (1.3 U/Umf)).Figure optionsDownload full-size imageDownload as PowerPoint slideHighlights► A defluidization time prediction model developed with influences of Al inhibition. ► Thermodynamics of the Al-Na-Si reaction system showed the agglomeration mechanism. ► An exponential decay of was found with the increase of the Al/Na ratio. ► The simulation data generally match well with the experimental results.