Based on studies on the chemical phase-analysis methods for rubidium in some mineral ores from Hunan Shizhuyuan mine, the occurrence state and mass fraction of rubidium in minerals were preliminarily determined by using inductively coupled plasma optical emission spectrometer (ICP-OES), combined with XRD, SEM, electron probe and MLA. After the phase separation of rubidium was determined and appropriate selective reagents and reaction conditions were selected, rubidium in mica, rubidium in fluorite, rubidium in silica-aluminum oxide, rubidium in feldspar, and rubidium in insoluble minerals were all analyzed. Then, a method for rubidium phase analysis was developed. It is found that this analysis method is characterized by lower detection limit, high precision, good reproducibility, wider linear range, and simple and rapid process. Based on an experiment on the interference of alkali metallic elements in minerals, reasonable analysis methods were selected according to different concentrations of K, Na and Li, so as to obtain accurate results with less interference.

A composite photocatalyst of TiO₂-CNTs was synthesized by adopting the sol-gel method and ultrasonic dispersion, and then its activity in the photocatalytic reduction of CO₂ was tested. With a load capacity of carbon nanotubes (CNTs) at 0.75%, the TiO₂-CNTs present a better performance in photocatalytic CO₂ reduction. Under UV irradiation with wavelength of 365 nm, the average yield of CH₄ and CO from the composite catalyst are enhanced by 90% and 156% respectively, compared to the usage of single TiO₂ catalyst. It is believed that the reaction is based on the following mechanisms: CNTs, as a co-catalyst, forms a compact electron transport structure with semiconductor of TiO₂, which significantly accelerates the transfer of photogenic carriers and reduces the electron-cave recombination efficiency. In addition, high specific surface area and excellent adsorption performance of CNTs can increase active sites for reaction, thus improving efficiency of photocatalytic reduction of CO₂ by the catalyst.

According to the characteristics of high-content aluminum and low-content lithium in hydrochloric acid leaching solution of a clay-type lithium ore in Hubei Province, lithium and aluminum were firstly separated by adopting calcination at 390 ℃ for 2.0 h. Then, the calcined residue, without grinding, was leached at 90 ℃ for 1 h with water in a volume ratio of 3∶1. After the leaching process, the leaching rate of lithium reached 93.85%, and no aluminum was detected in leaching solution, indicating a good Li-Al separation result. An analysis of chemical composition of calcined residue and leaching residue show that after water leaching, the main metal elements in the calcined residue, such as lithium, kalium, sodium, calcium and magnesium, all are leached into the lixivium, and the main composition of leaching residue are alumina and iron oxide.

Al-Cu-Mg-Ag alloy melt was treated with ultrasonic waves, and the effect of the ultrasonic melt treatment on the microstructure and hardness of alloy was investigated. The results show that compared to the as-cast alloy without ultrasonic treatment, the as-cast alloy after ultrasonic melt treatment for 90 s and 180 s respectively has its hardness correspondingly improved by 12.7% and 11.2%. The ultrasonic melt treatment can reduce the segregation of alloy composition and accelerate precipitation of Ω phase during 2 h aging process at 200 ℃. A quantitative analysis shows that ultrasonic melt treatment can reduce the grain size of as-cast alloy, but presents limited effect of fine grain strengthening. Solid solution strengthening and precipitation strengthening respectively improve the hardness of as-cast and T6-tempered alloys.

An experimental research was carried out on leaching of marine sedimentary manganese deposit from Xinjiang, and effects of acid-to-ore ratio, leaching temperature, leaching time, and ore particle size on manganese leaching rate were investigated. The results indicate that effective separation between elements of silicon and manganese in the solution can be actualized by increasing leaching temperature and prolonging leaching time, which can also benefit the formation of large-particle precipitates and enhancement of filtering speed. After manganese ores with particle size of -150 μm are leached at room temperature for 4 hours, with acid-to-ore ratio of 0.56, the manganese leaching rate can be up to 86.32% and a leaching residue rate is 61.12%, while the mass concentration of manganese ions in the manganese sulfate leachate is 41.29 g/L.

In order to explore the overall stability of slope before and after local failure, as well as the effect of excavation method on slope stability, the slope project on the north side of Jingxi-Barak mining area in Jinchuan of Xinjiang was taken to study local failure characteristics of slope and slope stability under different excavation state by adopting distinct element method. It is found that the distribution of the maximum main stress has a great impact on the slope stability, and the factor of safety for slope stability tends to decrease first and then increase from the stage before local failure to the stage after failure; with more excavated benches, the factor of safety for slope stability tends to increase first and then decrease; with the excavated benches more than 1, the overall factor of safety of the slope meets the requirements; with slope angle less than or equal to 60°, the overall factor of safety also meets the requirements, and the suitable slope angle is 60°.

Aiming at uncertainty of beneficiation product zone in dry magnetic separation process, an image segmentation method based on an improved U-Net model was proposed by employing machine vision. In this improved model, convolutional block attention module (CBAM) is utilized to enhance the recognition and attention of the network for target areas, which is beneficial to the segmentation of target objects under complex backgrounds; depth-wise separable convolution is adopted to reduce computational complexity while maintaining accuracy, providing strong support for obtaining high-resolution images of beneficiation product zone. Thus, this model can be applied in magnetic separation and also improve network performance. It is found that this improved model can bring segmentation accuracy up to 92.28%, and also is superior to classic U-Net, DeepLabV3+ and PSPNet models in terms of contour extraction completeness and denoising capabilities.

Manganese-rich phase with low impurity content was prepared with manganese sulfate crude solution by precipitation. A single-factor experiment was carried out to explore effects of reaction time, reaction temperature, pH value and feeding speed of precipitating agent on precipitation efficiency, and then the content of impurity in the finally prepared manganese-rich phase was also analyzed. After the leaching solution with Mn²⁺ in mass concentration of 38 g/L, and pH of 8.0 was precipitated at 60 ℃ for 4 h, with R as precipitating agent fed at a speed of 0.5 L/h, the conversion rate of Mn²⁺ was 94.3%. The ICP and EDS analyses showed that there were low content of impurity ions (K, Na, Ca, Mg) in the precipitate of manganese-rich phase.

An acoustic emission (AE) test of granite under uniaxial step loading was carried out, and the evolution characteristics of AE frequency, event incidence and fractal dimension of the sample at each load-holding stage before occurrence of instability and fracture were analyzed. The results show that the high-frequency signal first decreases and then increases with the increase of stress in the process of step loading; and macro-instability occurs in the sample at the last stage of step loading, at which the internal micro-cracks in the sample undergo from stable propagation to unstable propagation. The incidence of acoustic emission events decreases gradually as time is prolonged. At the last stage of step loading, the incidence of acoustic emission events decreases first and then increases, unstable fracture in the rock increases and micro-cracks gradually merge and coalesce to form macro-cracks, resulting in overall instability and fracture of the sample; with the increase of stress, the fractal dimension of acoustic emission decreases after an initial increase, and the rock fracture develops gradually in an order way from a disorder way.

For a cobalt leaching residue grading 1.31% Cu and 0.20% Co, a flowsheet consisting of grinding and a flotation process of one stage of roughing, one stage of scavenging and two stages of cleaning was proposed based on the occurrence of copper and cobalt therein. A closed-circuit test employing this processing technique can result in a bulk copper-cobalt concentrate grading 16.16% Cu and 2.06% Co with recoveries of 80.56% Cu and 68.29% Co, with the yield of 6.63%. It is concluded that with this approach, copper and cobalt resources in the cobalt leaching residue can be enriched and recovered efficiently.