UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fundamental study of the sedimentation of copper tailings with Kynch theory Li, Yuan


The management of tailings is a significant issue within the mining industry worldwide. An understanding of sedimentation and consolidation of tailings provides a solid foundation for the management of tailings. Kynch theory, the first mathematical formulation of sedimentation theory, is widely used as a tool in the analysis of batch sedimentation. However, due to various idealized assumptions within Kynch theory, the application of this theory to describe and analyze tailings sedimentation in the laboratory setting has not been well developed. For the purpose of this research, a new apparatus was designed and constructed to carry out column settling tests and to remove and collect samples of water and suspensions in thin layers. Tests conducted with this apparatus aim at investigating the settling behavior and the internal changes in the suspension during the process of sedimentation. Solid content and particle size were selected as the main parameters. Benchmark tests were first conducted to determine the general range of the two parameters. Solid contents of 35% and 40% were selected for the column settling tests. Tests were carried out on total copper tailings as well as fine portions of copper tailings with a particle size smaller than 37 µm. The experimental data collected from these tests was analyzed with Kynch theory. Total copper tailings, with a large range of particle sizes could not satisfy the assumptions of Kynch theory on particle size, and failed to conform to Kynch theory. In the tests of fines from copper tailings, the influence of segregation decreased as the coarse particles in the copper tailings were removed. Based on Kynch theory and some laboratory data, a prediction curve was developed and compared with the results of the column settling tests. Changes in the internal attractive force, and the influence of compaction settling were found to be reasons why the prediction curve was unable to completely match the experimental data.

Item Media

Item Citations and Data


Attribution-NonCommercial-NoDerivs 2.5 Canada