Search Documents

By A. Della Rocca, P. Jonker, B. Henning

Successful prediction of the collection efficiency of cyclone separators is a necessary step in cyclone design and verification. Aside from standard empirical relations, CFD modelling is widely adopted to supplement the design of cyclone units in relation to the desired separation performance. Nevertheless, effective adoption of numerical modelling is sometimes prevented by large modelling errors, especially for small size particles which are more subject to turbulence effects than larger particle classes. In the past, this motivated the adoption of more refined turbulence modelling approaches, such as large-eddy simulations. These approaches provide greater insight into time dependent behaviour, while leading to larger computational cost with respect to more standard approaches. In this paper, an effective Eulerian-Lagrangian Reynolds-averaged Navier-Stokes (RANS) approach is investigated, which is able to predict the complete separation efficiency curve with satisfactory accuracy, without requiring prohibitively long computational times. An assessment of the proposed approach is carried out against experimental data from large laboratory scale cyclones, in the form of mean and fluctuating velocity profiles from LDA measurements (Hoekstra, 2000), as well as with respect to the experimental separation curve (Zhao, 2005). The final application to a full-scale industrial cyclone unit is then carried out with the aim of tuning the cyclone design to the separation efficiency curve as required by downstream processes. Keywords: Cyclones, separation, CFD, RANS, RSM, Lagrangian

Jan 1, 2020

By L. Petzold

The demand for stainless steel is ever-increasing. As an essential element in stainless steel-making process, so is the demand of chromite concentrate. At the same time, the industry proven technologies for beneficiation of chromite are limited. Chromite is only crushed, screened and jigged to produce a concentrate. However, for ultrafine and low-grade material, no satisfactory beneficiation methods for the enrichment of chromite fines exist in industrial operation. This poses a clear challenge for the industry, which Primetals Technologies has addressed with the development of a new beneficiation process. The aim of this process was to enrich low grade chromite tailings to a quality suitable for the smelting process. Required concentrate grades are different over the world, concentrate grades of Cr2O3=40% and lower are found at the Bushveld Complex, South Africa. The target for the present study was a Cr2O3-grade in the area of 49%. Three different chromite tailings with Cr2O3 grades between 16 % and 29 % were part of a study, executed by Primetals Technologies. All three tailing materials were characterised by Mineral Liberation Analysis (MLA). Based on the physical, chemical and mineralogical findings, process flowsheets were developed. The process flowsheets included grinding to ensure a full liberation of the chromite. The liberated material was classified at a cut size of 63 μm to a coarse and fine fraction. Gravity separation and flotation processes were tested with the coarse particle fraction. Due to the limitations of gravity separation with fine particles, only the flotation process is suitable and was tested with the fine fraction < 63 μm. A Design of Experiment (DoE) with six factors, each three levels, was conducted to identify optimal parameters for the fine chromite flotation. Two different collectors were tested: a phosphoric acid ester and a fatty acid type collector. In small scale test work a chromite concentration suitable for the smelting process was achieved. On this basis Primetals Technologies will develop a pilot-scale test program to research recirculation effects and upscale potential of the process. Keywords: Mineral processing, tailings beneficiation, flotation, density separation, process development, chromite

Jan 1, 2020

By J. J. Cilliers, I. K. Craig, J. D. le Roux, K. Hadler

An analysis of the state-observability characteristics of a dynamic tank-to-tank model of a flotation bank indicates that the concentrate-flow rate of each tank can be estimated from accurate measurements of the pulp height in each tank and the final tailings flow-rate exiting the final tank. Therefore, there is an opportunity to construct an observer (model-based soft-sensor) which can estimate the concentrate flow-rate of tanks in a flotation bank. The concentrate flow-rate, i.e. mass pull, is generally measured by means of froth image analysers on industrial plants. A model-based soft-sensor will not only provide a means to validate froth image analyser results, but could potentially substitute them as they are less expensive to implement and maintain. Keywords: Concentrate flow-rate, estimation, flotation, modelling, observability, soft-sensor

Jan 1, 2020

By O. Dahl, B. Musuku, N. Schreithofer

Recovery of fine pentlandite is proving to be a challenging task in the Kevitsa sequential flotation operations, with over 60% of the nickel losses reporting into the -25μm size fraction. Since commissioning, the Kevitsa process has undergone major changes to address throughput and flotation performance. Major changes were implemented in the configuration of the milling and flotation circuits as well as in the operating conditions and reagent regime. Some of these changes improved nickel flotation while others deteriorated flotation performance. Most of the changes have had a direct impact on nickel flotation. This paper reviews the systematic analysis, laboratory testing and salient changes done to the process with special focus on the impact on the fine pentlandite flotation. Keywords: Blending, Flotation, Chalcopyrite, Pentlandite, Fines

Jan 1, 2020

By L. A. Balzan, M. Kalicinski, F. Nieuwenhuys

GEOSCANs on belt analysers have been widely utilised in iron ore for nearly two decades. Since the first installations in 2003, the availability of accurate and reliable grade data has become standardised across iron ore mines in the Northern Cape, where various streams are monitored continuously to ensure the optimal functioning of plants. Assmang Khumani has been at the forefront of technology adoption and has implemented 19 GEOSCANs at various locations throughout the plant, monitoring run-of-mine, stockyard feed, beneficiation plant feeds, and products. Over the years, collaborative advances in the technology have allowed for numerous benefits and gains to be made. In recent times, there has been a concerted effort on the development of analysis for phosphorus in the material flows, and based on Scantech's experience in the phosphate industry, this element has now been calibrated to achieve statistically acceptable and repeatable accuracies for real time analysis of this element on many of Khumani's GEOSCAN analysers. This paper provides an overview of GEOSCAN use at Khumani, while focusing on the development of phosphorus analysis. A case study of analysis results is presented for a range of analysers that have been calibrated for this element, as well as a discussion on the benefits obtained as a result of this analysis. Keywords: Online analysis, real-time, process control

Jan 1, 2020

By C. Chibwe, M. Tadie

Copper electrowinning models are useful for prediction of process performance, optimisation and improved understanding of complex phenomena such as current distribution without the expense of physical experiments. Recently, finite element models such as those generated in COMSOL Multiphysics have been developed for the electrowinning process. The predictive capacity of these models is improved by incorporating physical and chemical phenomena. Since physicochemical properties of electrolytes can have a significant effect on the quality of copper produced, this work was conducted to investigate the effect of such properties on electrodeposition at varying levels of copper, sulphuric acid concentration, iron concentration and temperature, in the presence of a smoothing agent. Tests were conducted at conditions similar to those for industrial copper electrowinning. Mathematical correlations were developed for the following properties; density, conductivity and diffusivity which were compared to existing models in literature and thereafter applied in a COMSOL generated model. The model predicts, current distribution at the cathode surface, deposit thickness (morphology), concentration of species within the cell and current efficiency. Current distribution output shows the impact of several phenomena on the deposit generated including the relationship between increasing copper concentration and current density and lack of influence of acid concentration on this property. Keywords: Electrowinning, physicochemical properties, electrolyte composition, modelling

Jan 1, 2020

By A. Ennen, R. Wagner, B. Orberger, J. -M. Milazzo, N. Uhlmann, J. Banchet, J. Lucic, N. Wörlein, M. Firsching, J. L. Dubos, C. Bauer

The residues from mining and processing ores can often be considered as secondary raw materials. However, the material is only available in the form of fine particles (< 3 mm). In order to recycle these wastes, it is necessary to agglomerate it for transportation prior to reprocessing. The resulting agglomerates need to mechanically resist the stress of transportation and loading into the furnace. X-ray computed tomography (CT) is an established method for characterisation of internal structures in material science. This non-destructive technique can provide information on size and distribution of particles and grains. However, the information is not limited to the size, but also can yield structural features like orientation and position of cracks and fractures, which are important features indicating potential for deteriorating the agglomerates. Furthermore, deduced features such as contact conditions of inclusions or boundary surfaces can also be calculated. We studied binder-free tablets of manganese-oxyhydroxide, bentonite, kaolinite and slag powder using micro-CT. Inclusions of highly absorbing particles were identified using a watershed transformation with automatic segmentation by Otsu thresholding. Subsequently, the resulting data was analysed regarding their volume percentage, distribution in size and location within the sample. For the slag agglomerates, the volume fraction of iron could be determined. Furthermore, visible cracks were investigated with the goal to determine the long-term durability. It was shown that the data provided by micro-CT is reliable and helpful for the characterisation of industrial waste agglomerates. Keywords: X-ray, computed tomography, micro-CT, agglomerates, manganese fines, slags, kaolinite, bentonite

Jan 1, 2020

By H. Lieberwirth, M. Klichowicz

The comminution at grain size level for the liberation of valuable minerals usually requires the highest size-specific energy. Therefore, the full understanding of the comminution process at this level is essential. Models based on the Discrete Element Method (DEM) could be a helpful tool for this purpose. In those, the mineral microstructure, which has a dominant influence on the fracture behaviour, can be described in great detail. One major concern, however, is the missing representativeness of simulated models. In this study, a method to overcome this limitation is presented. The authors show how a realistic microstructure can be implemented into a particle bed comminution simulation using a two-dimensional Bonded Particle Model (BPM). The improved algorithm based modelling approach is exemplarily compared to an equivalent real experiment. Thereby, the focus is set on the analysis of the interfacial breakage and the capability of the simulation to reproduce the experimental results. The comparison of the results shows that simulations at grain size level allow a more detailed evaluation of the fracturing behaviour. This is of special interest as the aim of comminution in mineral processing is not only the size reduction of coarse particles, but an efficient liberation of the valuable components. Hence, simulations with automatically generated real mineral microstructures will help to further improve the efficiency of ore processing. Keywords: Discrete Element Method, grain modelling, particle bed comminution, realistic synthetic

Jan 1, 2020

By Huanhuan SU, Jianqiang Zhang, Zhongyuan Liu, Junwei Ma

The high-sulfur bauxite in Henan Province was used as raw material. The grinding fineness, pH value and collector dosage on its desulfurization performance, and industrial flotation desulfurization process were studied. The best flotation desulfurization conditions were grinding fineness of 75%, pH of 8.5, and collector of 600 g/t. “One coarse, one cleaning and one sweep” closed-circuit flotation was established as the industrial flotation desulfurization process. After the closed circuit test, the sulfur content of aluminum concentrate with the yield of 93% was 0.23%, and the sulfur removal rate was 83.16%, and the sulfur content of tailings (sulfur concentrate) was 15.09%. Keywords：High-sulfur bauxite under coal mines, flotation, desulfurization

Jan 1, 2020

By A. J. H. Newell, K. Doherty

A copper-gold project has required a number of trade-off studies underpinned by test work to settle on pragmatic flowsheets and throughputs to support a pre-feasibility study that provides a compelling financial case. The primary trade-off studies were based on the mine schedule, ore type feed-grade recovery relationships, likely concentrate grades and typical marketing terms as well as the estimated capital and operating costs for a range of throughputs and mining strategies. Secondary trade-off studies were conducted for specific circuits and additional processing options. Although the details of these studies are not presented, the technical options and subsequent flowsheets are discussed. While the flowsheets will be optimised during the feasibility study stage, the basis is now fixed for the four ore types that would be mined from the copper-gold orebody. The ore types, resulting from a strong weathering profile, exhibit a wide range of copper mineralisation and gold associations, as well as gangue types. They display a range of characteristics that necessitated a nuanced approach to flowsheet development. The primary separation techniques that have been employed include gravity, classification, flotation and leaching while a range of comminution methods have been selected to suit the properties of the ore types. While technical issues are important factors in terms of maximising the quality of separations, recoveries and revenues, attention to material handling, equipment characteristics, the mine schedule, and capital and operating costs are equally important considerations for settling upon a suitable and robust flowsheet. Keywords: Flowsheet, testwork, studies, saprolite gold, native copper, chalcocite, chalcopyrite

Jan 1, 2020

By S. Krishnannair

Singular Spectrum Analysis (SSA) is an effective data-adaptive multimodal tool for monitoring chemical processes. However, monitoring multiscale signals using principal component analysis obtained from the multilevel decomposition of process signals using basic SSA assume the upper control limits of monitoring charts (" ) to have a Gaussian distribution. This assumption deteriorates the performance of basic SSA when applied to the monitoring of nonlinear processes. To address the issue in this paper, the kernel density estimation (KDE) was used to estimate upper control limits of monitoring charts for basic SSA based nonlinear process monitoring techniques. The monitoring performance of the resulting basic SSA-KDE approach was then compared with a basic SSA approach of which upper control limits of the monitoring charts were determined based on a Gaussian distribution. Application on nonlinear dynamic simulated process and base metal flotation plant chemical processes showed that the SSA-KDE approach provided better overall performance than the SSA approach with Gaussian assumption-based upper control limits for its monitoring charts. Keywords: Fault detection, singular value decomposition, singular spectrum analysis, kernel density estimation

Jan 1, 2020

By G. Bueno, L. M. Tavares, T. M. Campos

Iron ore pellet feed pressing has been one of the first and most successful applications of High- Pressure Grinding Rolls (HPGR) in the minerals industry. Vale S.A. has been one of the pioneers in this application, currently with six HPGRs in operation in the pelletising plants of the Complexo de Tubarão (Brazil). In spite of the important advances in the technology, risks still exist when scaling-up the process from small-scale information. The present work takes advantage of the recent advances in the mathematical modelling of the HPGR performance, using the population balance model to evaluate the modified Torres and Casali model prediction in iron ore pellet feed pressing for different scales of operation. Results are presented that compare the HPGR performance in different operational scales for iron ore pellet feed pressing, demonstrating the confidence of this model-based approach for scale-up to full-scale HPGRs used in pressing iron ore concentrates. Keywords: Scale-up, HPGR, pellet feed, modelling, simulation

Jan 1, 2020

By L. Forbes, G. Forbes, J. Wang

Bubble size measurement is a vital part of flotation system analysis and diagnostics. Over the years, a great many different bubble size analysis tools have been developed. However, all of them rely on ex-situ measurement and in-pulp sampling techniques. This paper evaluates the Mettler Toledo ParticleView™ Measurement probe as a novel method for in-situ bubble size measurement. This device is then compared to the conventional Anglo Platinum Bubble Sizer (StoneThree). It was found that the in-situ bubble size analysis technology appears to be a more user-friendly and reliable option for determining bubble size in flotation. However, the current instrument design is not suited to industrial applications. Keywords: Bubble size measurement, bubble coalescence, PVM probe

Jan 1, 2020

By S. Ramatsoma, D. N. C. Phillpotts, B. Whitehead

Changes in flotation performance have been correlated to changes in froth structure, velocity and stability. Literature has shown that when the aeration rate is varied for a specific float cell, the recovery of the valuable mineral reaches a peak which also coincides with a peak in the air recovery – a measure of froth stability. Peak air recovery optimisation methodology is a way to use direct readings of physical phenomena by froth cameras to optimise the metallurgical performance of a flotation plant. The aim of this paper is to show the transient nature of the peak air recovery value and the variety of conditions required to attain the peak air recovery on an industrial flotation plant with tight base-layer control of the feed and internal conditions. Industrial site tests that were conducted under controlled conditions showed different peak values at significantly different corresponding aeration rates during normal operation. The measured peak air recovery points corresponded to maximum PGM recovery during normal flotation, as expected. This dynamic nature of the peak air recovery point conditions has not previously been discussed in literature and indicates the need for online measuring equipment and informs the design of process control algorithms required to exploit the phenomenon for optimisation. Keywords: Froth flotation, flotation control, flotation optimisation, peak air recovery

Jan 1, 2020

By X. Chen, G. Wang, D. Wan

Computational fluid dynamics have been applied for detailed modelling of particle-bubble collisions. The most accurate approach in the previous work is using the point particle method to treat particles and bubbles. Nevertheless, the sizes of particles and bubbles are generally bigger than the Kolmogorov scale in the high turbulence of the flotation environment, which makes the point particle method inappropriate. This paper describes particle-resolved direct numerical simulation on the collisions between particles (denoted as heavy particles) and bubbles (denoted as light particles) in a homogeneous isotropic turbulent flow field, and a comparison with the point particle method.. A largescale stochastic forcing scheme was implemented within the mesoscopic multiple-relaxation-time LBM approach to maintain turbulence intensity at targetted levels. Both kinematic and dynamic collision kernels were evaluated to study the correlation of total collision events and turbulent kinetic energy (TKE) dissipation rate. The results show that collisions between particles increase with the TKE dissipation rate as radial relative velocities between particles increase with it. In the particle-resolved simulations, the dynamic collision kernels were found to not match with the kinematic collision kernels as in the point particle method. Quantitatively, they were approximately 4 to 10 times smaller than the corresponding dynamic collision kernels. Therefore, it is not reasonable to estimate the kinematic collision kernel in particle-resolved simulations as its rationales are not valid. However, the collision behaviour can still be accounted for by means of the dynamic collision kernel since it is obtained from the direct statistical analysis of the collision events. Keywords: Lattice Boltzmann Method; particle-resolved simulation; multiphase flow; collisions

Jan 1, 2020

By D. V. J. W. Groenewald, T. M. Louw, L. Auret, S. M. Bradshaw, A. V. Cherkaev

Unexpected equipment failures lead to financial losses due to production disruption. Time-based maintenance, based on checks over regular time periods, has been employed to minimise disruption. However, the improvements it can offer are limited as it may lead to unnecessary service of healthy equipment or missed failures occurring between service intervals. Condition-based maintenance (CBM) provides a further improvement to equipment reliability by detecting faults and predicting remaining useful life (RUL) based on equipment health monitoring. Its application in the mineral processing industry is hindered due to the high cost of health monitoring equipment. In this study, two stochastic process models, the adaptive auto-regressive model and the gamma process, are compared in their ability to forecast the degradation state and to provide RUL prediction. The gamma process model was found to be effective in both tasks. Due to its simplicity and minimum hyperparameters requirements, it is proposed to use the gamma process as a baseline for comparison with other models. Keywords: Condition-based maintenance, remaining useful life, slurry pump, adaptive autoregressive mode, gamma process

Jan 1, 2020

By Y. Fu, C. Aldrich

This study investigates the use of dense convolutional neural networks to estimate bubble size distributions in flotation froths. 20 videographically captured images of froths from an industrial platinum flotation plant in South Africa were labelled semantically with an image annotation tool, LabelMe (https://github.com/wkentaro/labelme). As a result, pixels in the images could be categorized into three classes, namely ‘bubbles’ (class 0), ‘bubble boundaries’ (class 1) and ‘background’ (class 2). A U-net architecture with a ResNet34 backbone was trained to segment the images via classification of the pixels in the images. 5 of the 20 images were used to validate the performance of the neural network. Explicit inclusion of the bubble boundaries as a class proved to be significantly more effective than training the network with two classes only (i.e. bubbles and background). However, the thickness of the boundary layer regions were critical. If too thin, it was difficult to distinguish between some contiguous bubbles and if too thick, some fine bubbles may not be identified. The results were promising, although there is still significant scope for improvement, among other by making use of more images for training and validation. Keywords: Flotation Froths, Convolutional Neural Networks, Deep Learning, Bubble size Analysis

Jan 1, 2020

By C. van Buuren, R. F. Sandenbergh, P. J. Pieterse

Gold extraction from double refractory ores using conventional cyanidation has proved challenging due to the presence of carbon within the ore. Carbon particles may adsorb leached gold resulting in gold losses to tailings, i.e. preg-robbing the gold. Some of the preg-robbed gold may be recovered by the application of the HiTeCC process on CIL tailings. The process developed by Metso Outotec involves caustic conditioning of the tailings followed by high temperature desorption and simultaneous adsorption onto virgin activated carbon of the aurodicyanide. The aim of this paper was to develop a model for the HiTeCC process which could be used to support design and optimisation of the process, based on the adsorption characteristics of the preg-robbing and activated carbons. It was found that accurate experimental isotherms can be used to predict the behaviour of preg-robbing adsorbents. Comparing the sorption isotherms for preg-robbing ores and activated carbon indicated similar dependencies on temperature and ionic composition. Applying boundary layer diffusion models coupled with experimentally determined isotherms for preg-robbing ore and activated carbon, the HiTeCC process was successfully simulated at experimental and plant scale. Gold recovery of the HiTeCC process is most sensitive to the temperature of desorption and gold loading of the activated carbon, with increased temperature improving recovery. The gold loading of the activated carbon in the process should always be maintained lower than that of the normalised pregrobbing carbon gold loading to ensure favourable conditions for adsorption onto the activated carbon. Caustic conditioning stabilises adsorption and is therefore not necessary during the desorption stages of the process, but potentially improves recovery during adsorption. Keywords: Preg-robbing, refractory gold ores, HiTeCC, aurodicyanide

Jan 1, 2020

By D. Minson, G. Lindner, S. Miskovic

Thickening is an essential process in the mining industry and currently one of the most costeffective technologies for recovering water from tailings streams. Improved performance of thickeners can lead to significant energy savings, and increased recovery of water close to the mill. Its performance is directly affected by the rheological properties of slurry at the raking zone. To further understand the dynamics of slurry inside thickeners, a multiphase particle-in-cell method (MP-PIC), is used to investigate its behaviour on a sliding channel. Parameters like particle-size distribution (PSD), rheological properties, and domain discretisation were benchmarked to study its influence on highly dense flows. The slurry concentration by weight of the samples ranged between 60 to 70% of solids by mass, under which conditions it is known to behave like a non-newtonian fluid. An experimental flume, 0.2 m wide and 2 m long, was used for validation under different angles and slurry composition. Visual analysis of the slurry height and shape, as well as velocity measurements, were compared with numerical data. The preliminary results show the ability of MP-PIC method to represent different types of highly dense flows, while still being computationally affordable and practical for industrial use. Keywords: High density thickener, CFD, MP-PIC, multiphase-flow

Jan 1, 2020

By A. Y. Senchenkoa, Y. V. Kulikov

The relationships revealed between ore hardness and specific energy consumed by crushing and grinding are described in the paper. A comparison of key technical and economical parameters of SAG and HPGR circuits is made, based on testing of several ore types for greenfield projects. The comparison of results make it possible to estimate efficiency of the technologies application, depending on ore hardness and taking into account potential risks for a circuit under design. Keywords: Comminution circuits, SAG, HPGR, test results

Jan 1, 2020