PhD thesis in mineral processing – Development of a predictive model of mineral concentration operations based on quantitative mineralogy

The development and dissemination of quantitative mineralogy techniques in the extractive industry offers numerous prospects for researching the potential of the entire mineral resource. It allows a detailed description of the material, from the very first exploration, at all stages of the development of a mining project and throughout the life of an operation until its rehabilitation. Although process modelling and simulation techniques are tried and tested and have long proved their worth, quantitative mineralogy brings new modelling perspectives and increased possibilities for the predictive nature of simulation. This combination makes it possible to address not only the main value products, but also co-products that meet market demand and the potential for reuse of production waste in localised applications, with a zero-waste perspective. This approach allows for the best possible orientation of laboratory or pilot plant tests that will validate the retained hypotheses while improving the implemented predictive models, models that are reused in the design and optimisation of the plant on an industrial scale.

Such an approach is still in its infancy and the few examples of application demonstrate a need for methodological clarification. CASPEO, which develops and publishes analysis tools for mineral processing, whether by statistical approach (measurement variability, data reconciliation by material balance) or based on modelling and simulation, has designed these tools with a view to managing and processing data from quantitative mineralogy. Although unit operation models are able to integrate them, improvements are still possible in order to better link mineralogy data to their performance.

As part of the development of our activities, we propose :

A CIFRE PhD position (M/W)

Orléans (France) 

Objective

• Main objective: development of a predictive model of mineral concentration operations (mainly flotation) based on a fine characterisation of the ores using, among others, quantitative mineralogy techniques in order to obtain sufficiently detailed data sets to best predict the valorisation of these ores by simulation.
• Secondary objectives: development of methods and techniques for the pre-processing of such characterisation data (such as discriminant analysis or parametric identification), the improvement of mathematical models of unit operations to take these data into account, and the application of the whole to the studied cases, such as in the framework of the EXCEED project.

Organisation of research work

Define methods and techniques for pre-processing data from quantitative mineralogy

• Determine the purely mineralogical data allowing the development and use of the different unit operation models which are considered.
• Identify raw measurements made by analytical instruments (such as QEMSCAN, MLA, Raman and IR spectroscopy, and even hyper-spectral imaging of cores) that could provide insights into the behaviour of materials and thus improve mathematical models.
• Define the additional analyses (chemical or other) that best complete the characterisation of the material, as well as the method for reconciling all these data.
• Estimate the accuracy of these measurements, both at the level of the analytical instruments and the sampling stages, and this automatically from the analytical results. Estimate the contribution of data reconciliation to increasing the accuracy of all analytical results.
• Identify the carrier phases of certain elements, such as radionuclides or other valuable (precious metals) or penalising elements, in order to be able to monitor their behaviour during treatment.

Relate quantitative mineralogy data to unit operation performance

• Analyse the different mineral liberation models with respect to liberation data from image analysis and complementary data from other analytical methods and processing tests. Determine their strengths and limitations with respect to multi-mineral liberation that may occur at different stages of the particle size reduction process.
• Propose a method of fine description of the material allowing a description of the multi-mineral liberation at the various stages of grinding that meets the needs of the modelling of unit operations and that can be fed by data from quantitative mineralogy.
• Improve grinding models to take into account this fine description.
• Determine laws, based on the physics of the phenomena or by statistical approach, linking the performance of mineral separation operations to this fine description by deconvoluting the effects of operating conditions from the separation potential linked to the material itself.
• Improve the various physical or physico-chemical separation models, mainly those of flotation, by introducing these laws while ensuring a prediction of the characteristics of the products in the form of the same fine description. The leaching stage, initial to the hydrometallurgical treatment, will also be addressed according to the same analysis criteria.

Validate the methods and tools described above for rare metal granite and LCT pegmatite deposits

• In the context of fine material characterisation projects using quantitative mineralogy, the PhD student will have access to the targeted ore samples. A back and forth between his work and that of the characterisation teams is essential, as the characterisation data will feed into the thesis work, the results of which will guide the characterisation techniques.
• The improved mathematical models will be used to build simulators for mineral processing plants for the deposits studied in the project. This also includes laboratory scale and pilot plant testing.

Propose a generalisation of the methods described above to other types of deposit, in the form of a “good practice guide”

• Describe the characterisation methods that can be used, those for the pre-processing of the obtained data
• Describe how to use them in a modelling and simulation approach, whether in plant design or optimisation.

Research project environment

This is an industrial research project being carried out at CASPEO, a company based in Orléans (France), as part of an industrial research training agreement (Cifre). The doctoral student will be a member of the company’s staff and will be based mainly at CASPEO’s premises in Orléans, with periods spent in ENSG’s laboratories, and even in other European laboratories.

This doctoral research will be jointly supervised by the University of Lorraine – Ecole Nationale Supérieure de Géologie de Nancy (ENSG), in the person of Professor Lev Filippov, and by CASPEO, in the person of Doctor Stéphane Brochot.