A combined approach of generalized additive model and bootstrap with small sample sets for fault diagnosis in fermentation process of glutamate

Evaluation of different factors on metal leaching from nickel tailings using generalized additive model (GAM)

Compared with sulfide tailings, the oxidation and transformation of certain substances in oxidized tailings into more soluble forms may affect the bioaccumulation and biomagnification properties and enhance the risk of toxic effects in the ecosystem. This study aimed to apply the generalized additive model (GAM) to evaluate factors affecting heavy metal leaching from nickel (Ni) tailings. We created an orthogonal experiment table (L₁₈(3⁷)) to evenly distribute the different combinations of factor values. The Ni tailings were immersed in solutions with different combinations of factor values for 16 d, and samples were taken on days 1, 2, 4, 7, 11, and 16 to measure the pH and heavy metal concentration of the leachate. The GAM was used to fit the concentration of heavy metals of the leachate and the initial factors of the leaching solution. The results showed that the pH and Cr concentration of the leachate increased with time and stabilized after 1 d (pH of approximately 7), while the Mn, Ni, and Tl concentrations gradually decreased and stabilized after peaking on the first day. An analysis of the GAM results showed that the Cr concentration was highly sensitive to the solid-liquid ratio (F = 127.8) and tailing particle size (F = 10.7). The Cr concentration of the leachate was significantly higher when exposed to a high solid-liquid ratio or a fine particle size, whereas the Mn, Ni, and Tl concentrations were highly sensitive to the KCl concentration and solid-liquid ratio (F = 77.4, 146.9, and 315.9 respectively). The GAM identified interactions between key factors, which have complex and strong effects on the leaching of tailings and the migration of heavy metals, either promotional or antagonistic. The prediction of the minimum Cr leaching concentration shows that GAM can be used to determine the conditions associated with minimum leaching concentrations of heavy metals and to effectively predict the metal concentrations of leachate. As such, the results of this study can be applied to the management of nickel tailings.