To explore green gold leaching reagents, a series of imidazolium cyanate ionic liquids (ILs), 1-ethyl-3-methyl-imidazolium cyanate ([C₂MIM][OCN]), 1-propyl-3-methyl-imidazolium cyanate ([C₃MIM][OCN]) and 1-butyl-3-methyl-imidazolcyanate ([C₄MIM][OCN]) were synthesized and characterized by Nuclear Magnetic Resonance (NMR), Fourier Transform Infrared Spectroscopy (FTIR) and thermogravimetric (TG) analysis. In this research, the imidazolium cyanates were utilized as a solute, which not only decreased the usage of ILs but also increased their gold dissolution capability. The gold dissolution performances of three imidazolium cyanates were characterized by dynamic leaching test and Scanning Electron Microscopy (SEM). The results show that the all three imidazolium cyanates had a gold dissolution ability, and the shorter the carbon chain on the imidazole ring in imidazolium cyanate, the faster the gold dissolution rate. The gold dissolution performance of [C₂MIM][OCN] was the best, and the weight loss of gold leaf was 2.9 mg/cm² at 40 °C after 120 h dissolution in [C₂MIM][OCN] mixed with 10 wt. % water. Besides this, the gold dissolution rate increased with the increase in the concentration of imidazolium cyanates as well as the reaction temperature. The gold dissolution performances of imidazolium cyanates in different solvents including water, acetonitrile, dimethyl sulfoxide (DMSO) and dimethylformamide (DMF) were also investigated, and the weaker the polarity of the solvent, the more conducive it was to the gold dissolution reaction. The mechanism of gold dissolution by imidazolium cyanates was investigated through NMR spectroscopy and Electrospray Ionization Mass Spectrometry (ESI-MS). It was inferred that during the process of gold dissolution, Au was oxidized to Au⁺ and the imidazolium cations were deprotonated to form N-heterocyclic carbenes, which coordinated with gold to form gold complexes and achieve gold dissolution.

In thiosulfate leaching of gold, the copper-ammonia complex serves as an oxidant and catalyst. This study examined the impact of magnetizing the copper-ammonia thiosulphate lixiviant solution on gold leaching from refractory oxidized ores. Magnetization reduced surface tension, improved wettability and infiltration, and enhanced the diffusion of leaching agents. It also increased dissolved oxygen content and boosted the catalytic efficiency of copper-ammonia complexes. These changes led to more efficient gold extraction, with column leaching showing a 4.74% improvement in extraction rates compared to non-magnetized methods and a 3.67% improvement over cyanide processes. These findings suggest that magnetized copper-ammonia thiosulphate lixiviant is a promising, environmentally friendly alternative to cyanide for refractory oxidized gold ores.

Accurate prediction of flotation concentrate grade is essential for maintaining product quality and supporting timely process adjustment. However, mechanistic prediction remains difficult because industrial flotation involves nonlinear multivariable interactions and partially observed operating states. Recent advances in machine learning have increased interest in data-driven methods, with encouraging results. Although recent studies have incorporated temporal dependencies into flotation-grade prediction, explicitly modeling the condition-dependent evolution of multivariate process–quality relationships under changing plant conditions remains challenging. In this study, industrial flotation records were systematically analyzed from a data-driven perspective to identify challenges affecting concentrate-grade prediction. Statistical analyses showed that variable–grade relationships are condition-dependent and vary across operating periods, indicating that measured variables only partially characterize evolving flotation states and that fixed input–output mappings may be inadequate. Accordingly, a DGraFormer-based model was developed to capture evolving inter-variable dependencies and multi-scale temporal patterns. For one-hour-ahead forecasting, the model achieved an RMSE of 0.7587 and an $R^2$ of 0.5423 for iron concentrate grade, and an RMSE of 0.6917 and an $R^2$ of 0.6384 for silica concentrate grade, outperforming the evaluated machine-learning and deep-learning baselines overall. These results underscore the importance of dynamic multivariate modeling for accurate concentrate-grade forecasting.