高岭石
煤浆
煤
泥浆
选择性
水煤
化学
化学工程
泡沫浮选
胺气处理
工艺工程
环境科学
矿物学
环境工程
有机化学
催化作用
工程类
作者
Liang Shen,Zihan Li,Jiabao Gong,Lingyun Liu,Erle Qiao,Yifang Liu,Fanfei Min
摘要
Abstract With the deterioration of underground coal quality, hundreds of millions of tons of coal slime in China need to be treated using flotation methods every year. However, the design and selection of coal slurry flotation collectors lack theoretical guidance. In this paper, the quantitative structure–activity relationship (QSAR) model between the molecular structure descriptors of collectors and the flotation recovery and selectivity index is established by using pure kaolinite/clean coal flotation experiments and molecular simulation techniques. Flotation results showed that the flotation recovery rate of kaolinite by collectors with larger head group size, such as DDMAC, DDP, and DTAC, is higher than that of the other collectors with smaller head group size. DDA, DDP, and DD have similar flotation capabilities for clean coal, and their selectivity was slightly better than others. The QSAR model between the molecular structure descriptor and the selectivity index was successfully constructed by combining flotation experiments and molecular simulation results. The reliability of this model has been proven through theoretical analysis and flotation experiments. The molecular simulation results indicate that all eight collectors can advertise on 001 (−) surface of kaolinite, but the final advertisement configuration is different. DDA has less coverage on the surface of kaolinite. When the collector is DDA, the concentration of water molecules near the 001 (−) surface of kaolinite is the highest, indicating that the collector DDA has the weakest effect on improving the hydrophobicity of the 001 (−) surface of kaolinite. DDMAC and DDBAC have the best facilitation to the surface hydrophobicity of kaolinite 001 (−). The research results have theoretical guidance significance for the selection and design of mineral flotation collectors.
科研通智能强力驱动
Strongly Powered by AbleSci AI