Use this URL to cite or link to this record in EThOS:
Title: Quantification in minerals processing via X-ray micro-tomography
Author: Reyes Leiva, Francisco
ISNI:       0000 0004 7658 6976
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Minerals processing has faced, over the last century, the challenge of an ever increasing demand for metals and at the same time the steady decline of head grades. To overcome this challenge metallurgists have developed new technologies to process the ore and new ways of quantifying the performance of mineral processing equipment. In this latter category image-based analyses have played an important role in making possible the link between performance and properties of the ore such as grade, mineralogy, grain size, porosity, exposure and mineral associations. Two important image analysis techniques are X-ray Micro-Tomography (XMT) and Automated Mineralogy based on Scanning Electron Microscopy and Energy-Dispersive X-ray Spectroscopy (SEM/EDS). The first has the benefit of being able to delineate the internal structure of the ore in three dimensions in a non-invasive and non-destructive manner. The second, conversely, is destructive and invasive but offers better resolution and direct mineral identification, although only in two dimensions. It is, thus, one of the objectives of this thesis to formulate a methodology that allows the combined use of these imaging modalities in order to produce a more complete quantification of the ores. The shared information opens the possibility to understand the level of uncertainty in the quantification of the ore's properties such as grain size and stereological effects. Another benefit of this methodology is that XMT scans can now be calibrated so that mineral identification can be performed with an understanding of the level of misclassification. The result is thus a more complete assessment of the ore's properties compared to the traditional individual assessment. The benefits of this new assessment methodology are shown for two important minerals processing systems, heap leaching and liberation by comminution. Heap leaching performance is the result of a complicated interaction between fluid flow, mass transport and chemical reactions, with the texture of the minerals having a strong influence on he resultant apparent leach kinetics. The prediction and simulation of these processes is thus complex and reliant on the availability of appropriate input data. This thesis presents a method for measuring the kinetics for the leaching of individual mineral grains, with the dependency on mineral type, grain size and distance to the particle's surface being assessed. Importantly, the variability in the kinetics was also measured, with this being an important factor in the time evolution of the apparent leach kinetics. The XMT images of the particles, together with the kinetic variability, were then used as the basis for a parallel particle scale leach simulator. This simulator is able to predict the long term leach rate of different sulphide species based on the initial leach behaviour, a useful tool for rapid leach assessment. Liberation has been traditionally studied using Automated Mineralogy systems, and has thus being restricted to two-dimensions, with the associated stereology related impacts on the data. In this thesis the benefits of using a three-dimensional assessment are shown. For the first time, mineral grade, surface exposure and particle size have been directly related to each other for a real ore. This allows for substantially improved quantification of mineral liberation and preferential breakage. The quantification of liberation and its association with surface exposure is an important consideration in froth flotation performance. The restricted sample size and stereological biases associated with two-dimensional SEM/EDS analysis has made accurate quantification of this relationship hard. In this thesis a simple model for this relationship is presented. The future direction of this work will thus be to develop improved models for predicting the relationship between ore texture, comminution technique and the flotation performance.
Supervisor: Neethling, Stephen J. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral