Sunday, 23 June 2019

Report on Physical Separation '19

Physical Separation '19 was the 6th in the physical separation series of MEI Conferences, and was held for the first time at the National Maritime Museum in Falmouth, Cornwall. The 2-day event, from June 13-14 2019, was sponsored by Holman-Wilfley, Steinert, TOMRA Sorting Solutions, Outotec, Master Magnets, Hudbay Minerals and CRC ORE. Media partners were International Mining and Imformed and industry advocate was CEEC. The conference was attended by 83 delegates from 18 countries.

This posting will summarise the papers presented at the event, social events and people being highlighted in the postings of 14th June and 15th June.
Thursday June 13th

For almost 40 years engineers have explored opportunities for using microwave energy to improve the efficiency of mineral and metallurgical processes.  Given the vast energy consumption of such processes this is not a surprise  as microwave heating has long been assumed to reduce energy consumption in process engineering unit operations.  Selective heating of microwave-absorbent sulphides and metal oxides deported in a microwave-transparent gangue matrix results in differential thermal expansion of the heated phase, yielding micro-fracture around grain margins. Subsequent downstream processing may then yield higher recovery of valuable mineral sulphides and/or lower specific comminution energy, compared to non-microwave treated ore.
But while the mechanistic principles are well established, the scientific and engineering challenges of developing a commercial scale system have been immense. Typical throughputs of a large copper mine can be in excess of 5000 tph of milled ore and a microwave based treatment system would need to handle equivalent throughputs. This is at least an order of magnitude higher than any other microwave process ever built.
However, recent work has shown for the first time that microwave technology can be used in commercial mineral processing plants and that it can be used at significant scale with several of the largest microwave processing plants ever built being applied.  A multidisciplinary team of engineers from the University of Nottingham and Teledyne e2v have developed a revolutionary process for extracting copper from its ore by exposing rocks to powerful microwave energy for a fraction of a second. The team combined their skills in microwave technology and engineering to develop the largest microwave processing system ever constructed, capable of processing up to 3000 tonnes of ore per hour.
We were fortunate to have the team leader, Prof. Sam Kingman, deliver a keynote lecture "What's cooking in mining?" which examined the steps required to scale up such processes, and to draw conclusions as to the steps required to see this technology in daily use across our industry - a time which he feels may be sooner than some workers may have previously thought.

Sam Kingman (right) with Dan Alexander

Canada's University of Toronto is also very much involved with the commercialisation of microwave technology and three papers were presented this morning on the subject. One of the barriers to scale-up has, according to Erin Bobicki, been a lack of understanding of how microwaves interact with minerals and ores. Key parameters include the real and imaginary permittivities, which influence both the heating rate and microwave penetration depth. Over the past years, the University of Toronto and Queen's University have conducted high-temperature permittivity measurements on various materials using the cavity perturbation technique. Erin presented permittivity data as a function of temperature across a range of frequencies for a variety of minerals and ores. On the basis of this data, recommendations were made for how microwave technology can be best utilised for sustainable metallurgical processing. 
In the second paper from the University of Toronto, Darryel Boucher showed why, although the mining of ultramafic ores is required to meet society’s continued demand for nickel, these deposits are difficult to process due to their serpentine content. Serpentine increases slurry viscosity, slimes-coats the valuable Ni-bearing mineral pentlandite, and reports to and dilutes the froth during flotation via the formation of bubble-fibre aggregates and entrainment. Darryel showed how microwave pre-treatment can be used to convert the serpentine gangue to olivine, which should improve ore processing. 

Erin Bobicki and Darryel Boucher with Kai Bartram of Steinert

Also on the subject of nickel, Toronto's Matthew Goldbaum discussed why, due to its low concentration of nickel and high sulphur content, the current practice is for pyrrhotite to be rejected from conventional nickel sulphide processing circuits. However, the nickel content of pyrrhotite tailings (up to 1 wt.% Ni) represents a significant resource, and a new flowsheet, based on the microwave heating of the pyrrhotite tailings to produce FeNi and FeS phases is under development. 

Matthew Goldbaum with Carolina Vargas of TOMRA
Improvements in grindability and energy consumption using microwave pre-treatment have been reported extensively in the literature. However, as argued by Ahmad Hassanzadeh, of the Helmholtz-Institute Freiberg for Resource Technology, Germany, the impact of the location of the microwave irradiation in a communition circuit has not yet been addressed. Collaborative work between Germany and Iran has assessed the influence of microwave location, exposure time and grinding time on particle size distribution, particle liberation degree and energy consumption for a porphyry copper complex deposit, showing that the correct location of the irradiation can significantly improve the grindability of materials.
There are a number of potential technology opportunities to effect up-front gangue rejection in the minerals sector. Australia's CRC ORE, one of the conference sponsors, has been working with several of its mining participants to assess the efficacy of a range of such technologies, across a variety of operations and commodities. CRC ORE's “Grade Engineering®” approach references a suite of integrated technologies across several separation “levers” – dependent on the textural amenability of the ore domains concerned.
Ben Adair, CEO and Managing Director of CRC ORE, detailed recent implementation trials of Grade Engineering® at the Sumitomo-owned Minera San Cristobal operation in Bolivia. This Pb/Zn/Ag mine operates with some of the lowest head grades in the industry – yet ranks as one of the most profitable. The implementation strategies at the site trials were presented, the upgrading of “mineralised waste halos” to ores detailed, and the textural attributes that facilitates gangue rejection at scale. This success has implications for unlocking “hidden value” across the sector. 

Ben Adair with Dan Alexander

Following on from this, Raoni Lage, of the JKMRC, Australia showed that mineralisation is not evenly distributed across the whole particle size distribution after blasting and crushing. For some ores, valuable minerals can concentrate in the finer size fractions. This grade by size deportment has been utilised by the Grade Engineering™ (GE) approach, which results in coarse gangue rejection. Raoni described a study in which the effect on coarse gangue rejection by using precisely controlled impact breakage energy applied to single and multiple ore particles was investigated. The JKMRC Rotary Breakage Tester (JKRBT) and a lab-scale Vertical Shaft Impactor (VSI) were used for breaking individual particles (particle by particle) and high feed throughputs (multiple particles) respectively. The amount of target metal in the coarse progenies was compared to investigate the benefit of particle interaction within a lab-scale VSI for coarse gangue rejection. 

Raoni Lage (centre) with Bernard Agbenuvor and Kevin Galvin

Particle breakage by impact was also the subject of the next two papers.  Gregor Borg, of PMS GmbH, Germany and Martin Luther University Halle-Wittenberg, Germany, showed how the innovative VeRo Liberator® applies a mechanical high-velocity comminution principle, where numerous hammer tools rotate clockwise and anti-clockwise on three levels around a vertical shaft-in-shaft (hollow shaft) system. The resulting high-frequency, high-velocity impacts cause a highly turbulent particle flow and trigger fracture nucleation and fracture propagation preferentially at and along mineral boundaries. Breakage of coarser particles occurs from the high-velocity stimulation of bulk ore particles, where the elasticity and compressibility modules control differential particle behaviour. The improved breakage behaviour results in reduced energy consumption and very high degrees of particle liberation in the relatively coarse fraction of the product.
On an industrial scale, Anglo American applies already two VeRo Liberators® in their South African operations and Penoles and Fresnillo, Mexico, have carried out advanced test-comminution of ores from several of their gold-silver-zinc-lead mines in Mexico. Gregor described results from test-comminution by VeRo Liberator® of ores from Cienega Mine, Durango Province, Mexico, in comparison to Cienega’s conventional comminution by SAG milling and ball milling. The results show that the VeRo Liberator® achieves a similarly high degree of particle liberation compared to the SAG mill, but at a drastically coarser particle size. 

Gregor Borg (right) with Oscar Scharfe
In a further JKMRC paper, Bernard Agbenuvor showed how precise impact breakage has been identified to have the potential to enhance early coarse gangue rejection. This form of breakage targets to concentrate valuable minerals into finer fractions, which allow utilising a simple screening for rejection of coarse economically barren material.  Bernard described how the effect on coarse gangue rejection by using single and multi-stage precise impact breakage techniques was investigated. It was found that subsequent stages of precise breakage reduce the amount of valuable minerals in coarse particles and has the potential to improve the outcome of early coarse gangue rejection. 
In the final paper of the morning Daniel Parvaz, of SELFRAG AG, Switzerland, described how a continuous electric pulse fragmentation system has been used by the SAIDEF incinerator in Switzerland since 2016. This system is the first of its kind and has led to many additional developments in the processing of incinerator bottom ash (IBA). Disposal of IBA, the solid residues of municipal waste incineration is a challenge, as the metal content can be up to 12%, classing it as a ‘reactive’ waste, increasing landfill costs. Metal recovery decontaminates and reduces the volume to landfill and generates additional revenue.

The afternoon session was dominated by eight papers on ore sorting. With the depletion of global ore deposits, there has been an increasing industry trend towards mining of lower average grade deposits in order to meet global demand, resulting in increased processing cost, energy usage and water usage. Consequently, there is significant interest in pre-concentration technologies, and bulk ore sorting is one approach that is gaining increasing attention.
As pre-concentration at crushing sizes is dependent on the ore texture and consequently relies on the mineral liberation degree at those sizes, a straightforward methodology to assess mineralogical quantitative data of the ore macro texture has been developed, and was discussed by Rui Sousa, of the National Laboratory of Energy & Geology, and the University of Porto, Portugal.

Daniel Parvaz and Rui Sousa

Nick Cutmore, of CSIRO, Australia, described the development of a large scale magnetic resonance based technology for the rapid quantitative detection of selected copper, iron and arsenic minerals in bulk ore. The technology is suited to measurement on large primary conveyors for enabling “pod” based bulk ore sorting, where barren ore pods, at the 1-10 tonne scale, are diverted from conventional processing. The technology has been demonstrated for a wide range of copper, arsenic and iron minerals and full scale ore sensing and sorting systems are under trial.
All round the world the processing of coal, ores and industrial minerals is combined with the use of a huge quantity of water. This water must be obtained and - after use - recycled, discharged, or it leaves the process with the product or the tailings. This generates quite large environmental challenges. To overcome this, Jens-Michael Bergmann, of conference sponsor TOMRA Sorting, Germany, described how TOMRA developed its own, unique X-ray transmission (XRT) sorting technology back in 2003, using the patented Dual Energy XRT approach. The TOMRA XRT-technology enables materials to be recognized and separated based on their average atomic density in a water-free way. The technology can create an extremely high purity level in the sorted materials irrespectively of size, moisture or contamination. In recent years TOMRA has gained much experience applying this technology in the recycling industry and on various minerals and ores. The sorting performance and the feed capacity have increased considerably; nowadays one modern sorting machine can (depending on grain size) handle up to 750 tph feed capacity and accurately remove the unwanted contaminants.

Jens-Michael Bergmann (2nd left) with Carolina Vargas, Jeff Gurtler and John Williams

James McFarlane, of Mining Plus UK Ltd, also acknowledged that ore sorting technology has proven and well understood downstream benefits on a mining and processing operation. However, the upstream impacts are often less well appreciated and offer opportunities to maximise the economic impact on an operation. James presented a study on work undertaken by Mining Plus at the Hemerdon tungsten/tin deposit, which was exploited briefly by the ill-fated Drakelands Mine, utilising the results of an on-site Steinert XRT ore sorter pilot plant to maximise the economic impact across the mining operation. The approach capitalised on the geo-metallurgical block model developed for the project and results showed improved project economics in all ore sorting scenarios, which were used to develop a revised grade control system with impacts on mining method, fleet size, ROM pad design and layout, mill feed strategy and production forecasting.
The performance of a sensor-based sorting operation is affected by mechanical and physical factors along with the correct identification of the materials. Hence, it is necessary to ensure and monitor that all the accurately identified particles are reported to the correct product stream, either “accept” or “reject”. Ergin Gülcan, of Hacettepe University, Turkey, presented a novel approach for sensor based sorting performance determination and compared the results with previous approaches.
Jacek Kolacz, of Comex Polska, Poland, introduced a new sorting system which has been developed by Comex to achieve efficient separation at low mineral content levels. The system employs high sensitivity X-ray sensors providing multi-energy analysis, thus giving maximal information about processed materials. This allows analysis of the internal structure of the particles and to directly detect intrusions of minerals and metals like copper or gold, while allowing high efficiency analysis, quantification and removal of rock particles representing a waste fraction with very low metal content.

Jacek Kolacz (right) with Erik Larsen
Aaron Young, of the University of Utah, USA, described collaborative work between Brazil, Spain, Poland and USA on the affinity of Dual Energy X-ray Transmission (DE-XRT) sensors on minerals bearing heavy rare earth elements. DE-XRT sensor affinity was evaluated for forty-two xenotime heavy rare earth element (HREE) grab samples from the Madeira/AM deposit. In a paper from the Helmholtz Institute Freiberg for Resource Technology, Germany, Laura Tusa then showed how the potential of hyperspectral short-wave infrared (SWIR) sensors combined with machine-learning (ML) routines have been exploited to improve sorting performance. The approach adopted can be adjusted to optimize sorting for a variety of ore types and thus could increase the attraction of sensor sorting in the minerals industry.
 In the final paper of the day Bertil Pålsson, of Luleå University of Technology, Sweden, described the use of electrically heated screen decks for producing narrow sub-millimetre fractions. He showed how the probalistic screen may be evaluated in the same way as a classifier with a cut point and sharpness of separation for each screen deck, and as a total for combinations of screen decks.
It was a great first day at the conference but nice to get some fresh air and in the late afternoon we met up on the seafront for a rather damp 3 mile walk around the Fal estuary to the inner harbour and drinks at the very crowded but atmospheric 17th century Chain Locker pub.

A wet walk by the Fal estuary
Welcome drinks in the Chain Locker
Friday June 14th

Hydrocyclones, the devices most commonly used for classification in the mining industry, were the subject of the first four papers of the morning.
Even though they are in very common use, hydrocyclones don’t get as much attention as more expensive comminution units, and their technological progress has been incremental, according to Vladimir Jokovic, of the JKMRC, Australia. The SMI-JKMRC has conducted a classification study using semi-inverted hydrocyclones. The results from the study were encouraging with lower water split to underflow (WS) and much coarser cut sizes. A hydrocyclone with lower water split to underflow (WS) will improve classification of fines which would otherwise be sent back to the grinding circuit. The ability of hydrocyclones to cut coarser would potentially expand their application in the areas where fine screens are currently in use. Therefore, the very low WS and coarser cut are of paramount importance for a step change in future industrial operation. Vladimir presented the classification performances of semi-inverted hydrocyclones in his paper.
Mini-hydrocyclones have been applied successfully to micron-size particle separation since their diameter is directly proportional to the cutsize that can be achieved. Their use for the processing of fine (less than 20 µm) ore requires a better understanding of how design and operating parameters affect separation performance. In particular, the length of the mini-hydrocyclone’s conical sections has not received much attention, arguably due to the lack of commercially available units of different heights. Pablo Brito-Parada, of Imperial College, UK, showed how this limitation has been overcome by using 3D printing to manufacture specific designs and assessing conical heights along with feed flowrate and solids concentration in a CCRD experimental design. Good particle classification was achieved, which was not affected by design or operating conditions. A linear effect of conical height and feed flowrate on recovery was observed. Interestingly, conical height exhibited a quadratic effect on concentration ratio, thus making this a critical variable to consider for clarification/thickening applications.  

Pablo Brito-Parada (right) with Christian Ihle

Alireza Eslamian, of ESS Engineering Software Steyr GmbH, Austria, presented work on a novel coupled DEM-SPH method for simulation of complex fluid flows inside hydrocyclones. The simulation of slurry flows that carry multidisperse particles is a vital issue for optimal design of hydrocyclones which are used in different mineral material processing lines, especially in dense media separation technology. Although various CFD-DEM methods for simulations of the two-way coupled interaction of particles with slurry flows inside hydrocyclones have already been proposed, almost all of them suffer from high computational cost problems. 

Alireza Eslamian (right) during the coffee break
Process modelling packages used for purposes such as process design and control of mill circuits generally use empirical and semi-empirical correlations to determine the performance of each of the unit operations in the circuit.  It is well known that each of these correlations is subject to uncertainty. Phil Schwarz, of BGRIMM, Australia,  focused on correlations that are presently used for hydrocyclones. Computational Fluid Dynamics (CFD) modelling holds promise for predicting performance of hydrocyclones, but such modelling is also subject to uncertainties, particularly when the solids concentration is significant; however, though the predicted values may not be accurate, the dependence of predictions on operating conditions should be more reliable. Phil showed how the results of CFD simulations are used to determine the dependence of the primary performance measures on geometry and operating conditions.  Performance measures analysed were cut-size, pressure drop, water split and sharpness of cut, and the dependencies compared with those given by the Nageswararao model and the Flintoff-Plitt model. The results point the way to a methodology in which the best aspects of experiment, CFD simulation and process modelling can be combined.
Phil Schwarz with Yann Foucaud
Gravity separation then dominated the rest of the morning, with Patrick Zhang, of the Florida Industrial and Phosphate Research Institute, USA, showing how recent technological breakthroughs have paved the way for commercializing an innovative gravity separation device, the packed column jig (PCJ). Partick presented three case studies of PCJ use for processing tailings/byproducts from the phosphate industry. The world’s phosphate industry generates over 150 million tons of phosphogypsum (PG) annually. The phosphate companies are under great pressure to find uses for PG quickly and the use of PG as construction material could reduce PG accumulation dramatically. A combination of gravity separation using PCJ with flotation produced an upgraded PG acceptable for many construction uses, with soluble P reduced from 0.89% to 0.04% and whiteness increased from 39 to 87.

Patrick Zhang (centre) in one of the technical sessions
Significant levels of ultrafine particles remain in the feed after desliming in cyclones. These slimes remain a challenge for downstream processing due to the increased viscosity, moisture retention in the product including associated materials handling issues, and entrainment of contaminant with the product. James Carpenter, of the University of Newcastle, Australia, examined the influence of slimes on the recovery of high-grade iron ore fines across a size range of 0 – 0.3 mm using the REFLUX™ Classifier. Experiments were performed using closely spaced inclined plates (3 mm apart and 1000 mm long) over a range of solids throughputs of 3.5 – 28 t/(m2 h). Lower feed solids concentrations led to higher recoveries due to the reduction in the viscosity arising from the slimes. At the higher volumetric flowrates used to increase the solids throughput, the very high shear rates generated in the channels further reduced the viscous effects of the slimes. The results are supported by rheological data, based on the ultrafine solids collected from the overflow stream.
After the coffee break the inventor of the REFLUX™ Classifier, Kevin Galvin, of University of Newcastle, Australia, showed how a Reflux Classifier was used to recover zircon from an ultrafine, low grade mineral sands feed below 300 microns, deslimed to varying degrees below 38 microns. This system consists of a fluidized bed with a set of inclined channels above. The work builds upon previous studies concerned with mineral sands, revealing the benefits of using an inclined channel spacing of 3 mm. 

Kevin Galvin (right) with Stuart Saich
Karlo Leandro Baladad, of the University of the Philippines, presented a study investigating the effect of electrostatic interactions in the separation of magnetite and silica in aqueous media using enhanced gravity concentration. Magnetite and silica suspensions at different coagulation conditions were prepared and subjected to enhanced gravity concentration using a Falcon L40 concentrator. Increased recovery of both magnetite and silica was observed at pH 3 where attractive forces are strongest both for one-component and two-component interactions. However, the increased recovery of both particles caused a decrease in separation efficiency at this condition. At pH 11, where repulsive forces are dominant for both one-component and two-component interactions, recovery of both minerals decreased. The highest separation efficiency of 64.1% was at pH 9. At this condition, two-component interactions were repulsive which led to the increased removal of silica. 

Karlo Baladad (right) with Jane Mulenshi, Amanda Wills and Jon Wills
Scheelite is one of the main minerals for the production of tungsten. It is usually encountered in skarn deposits where it is commonly associated with other calcium minerals as fluorite, apatite and calcium silicates. Worldwide scheelite is beneficiated into chemical grades by direct flotation, but the separation efficiency remains limited due to the similar flotation behaviours of scheelite and gangue minerals with fatty acid. The only solutions used to overcome this issue involve high energy consumption or ecotoxic reagents. Yann Foucaud, of Université de Lorraine, France, described a novel method based on the use of a centrifugal Falcon concentrator to perform an efficient elimination of gangue minerals and fine particles as well as an acceptable scheelite recovery, enabling a decrease in flotation reagent consumption. 
Carl Bergmann, of Mintek, South Africa, showed the value of incorporating mineralogical data into modelling the concentration of chromite ores using gravity spirals. The approach used is based on mapping the deportment of many thousands of individual particles across rougher, cleaner and scavenger spirals at varying product yields. AutoSEM characterisation, using a Mineral Liberation Analyser (MLA), was used to reconcile mineral deportment to measured size and assay data.  Results from 4 different chromite ores from the Bushveld Complex were compared using partition surface models to simulate spiral performance. The benefit of this approach is that the effect of varying head grade, mineral liberation and particle size distribution on circuit performance can be evaluated.
A novel system involving a gas-solid fluidized bed subjected to vibration was described by Deepak Kumar, of the University of Newcastle, Australia. A mesh with an aperture of 1 mm, incorporating a large central hole referred to as a Sink-Hole, was located just below the surface of a fluidized granular media. The fluidized granular media, nominally 0.3 mm sand particles, expanded through the large Sink Hole, spilling onto the top of the mesh, and then sifted back into the lower fluidized zone. The effective density of the medium at the mouth of the Sink Hole was determined by examining the probability of a large tracer particle of a given density floating as opposed to sinking. The effective density was deemed to be the condition denoting a 50 % probability of the tracer particle passing through. Interestingly, this density condition greatly exceeded the density of the medium and invariably the density of the granular media itself. Moreover, the effective density was largely independent of the size of the tracer particles covering the range 2.8 to 8.0 mm, a sign of genuine density-based separation. 
In the final paper of the morning Ozan Kökkiliç, of McGill University, Canada, highlighted how, due to water shortages, developing dry separation methods to replace the commonly used water-based separation has become crucial. One of the candidates for dry gravity processing are gas-solid fluidized beds. The behaviour of bed and solid motion in a fluidized bed have been investigated over the past few decades using various visual and numerical methods, but there are not enough studies focused on the separation behaviour of the fine particles. Ozan detailed the investigation of particle motion through a fluidized bed using the positron emission particle tracking (PEPT) technique. This single particle tracking method is a powerful way to provide knowledge about the separation mechanisms through direct visualisation of the particle trajectory, determined from records of the particle position over time. In this study, by tracking an activated quartz particle, the movements of different sized beds were characterized and then by tracking an activated hematite particle the separation of dense particles in a fluidised bed was observed for different size fractions.

The afternoon session was dominated by case studies.
The Nechalacho rare-earth element (REE) deposit is located in the Northwest Territories, Canada. The main rare-earth minerals (REM) in the deposit are zircon, allanite, bastnäsite, synchysite, monazite, columbite (Fe) and fergusonite; with quartz, feldspars and iron oxides accounting for most of the gangue. Recent studies have demonstrated the potential for early gangue rejection, of a relatively coarse feed, using a spiral concentrator, prior to fine particle gravity separation using a Multi Gravity Separator (MGS) to produce a concentrate which comprises primarily REM and iron oxides. These studies were, however, limited in scope, with each technique tested separately and only demonstrated a proof of concept. Ozan Kökkiliç described expansion of this work by investigating a complete flowsheet incorporating early gangue rejection at coarse particle sizes, followed by fine grinding of the pre-concentrate and fine particle gravity separation. The final gravity concentrate is then further upgraded using magnetic separation to separate the valuable REM from the iron oxide gangue.
Currently, about 75% of the EU's chromium is being imported. In the value chain of stainless steel production, significant amounts of chromium are lost in ferrochromium (FeCr) slags and stainless steel (SS) slags. In a slag chromium is mostly present in entrapped metallic particles (Fe-Cr alloys) and/or stable spinels.  To unlock the potential of these low-grade resources, a new approach to metal recovery must be deployed. Fero Kukurugya, of VITO - Flemish Institute for Technological Research, Belgium, described European H2020 project CHROMIC, which aims to develop a process for recovering chromium from SS and FeCr slags based on the smart integration of physical separation and subsequent pyro- or hydrometallurgical treatment. Results of physical separation show significant Cr up-concentration by using magnetic separation techniques and density separation (wet shaking table), where Cr content can be increased almost by a factor of 4.   
Nickel is an important metal with a total global consumption of 2 million tons per year. It is sourced from sulphide and laterite ores, but currently laterites are becoming more attractive for nickel production due to the depletion of high grade nickel sulfide ores. Laterites are generally processed by hydro or pyro-metallurgy and therefore, pre-concentration of nickel before such processes is important. About 60% of land-based nickel reserves are present in laterites with nickel and cobalt grades of 0.8 to 3% and 0.05 to 0.15%, respectively.
Saeed Farrokhpay, of Université de Lorraine, France, discussed the effect of different physical methods on the pre-concentration of nickel in laterite ores. The ores were sourced from different geological zones (saprolite, limonite and intermediate zone) of a laterite mine in South East Asia. Nickel was upgraded using physical methods such as magnetic separation, gravity separation, and sizing. The effect of ore mineralogy on choosing the best strategy to upgrade nickel was highlighted, as was upgrading nickel using a combination of these methods. The findings of this project will help to unlock a substantial volume of nickel with significant value from laterite ores.
An alternative process for laterite treatment, combining pyro-metallurgical pre-treatment followed by mechanical processing to decrease the feed for a subsequent rotary kiln electric furnace (RKEF) process, was described by Klaus Hahn of RWTH Aachen University, Germany. The pre-treatment process consisted of a non-selective segregation of contained metal phases, which changes metal properties and transforms nickel, cobalt and iron to metal alloys with grades around 80% for Ni and Co. Due to its fine grain sizes below 20µm, subsequent processing involving fine grinding and magnetic separation was investigated.
Repositories of historical tailings pose environmental risks but could also become metals and minerals resources, as relatively high mineral and metal contents are due to less efficient extraction methods in the past and/or relatively low metal prices at the time. Jane Mulenshi, of Luleå University of Technology, Sweden, described how the Smaltjärnen tailings repository in Yxsjöberg, Sweden was sampled and the main layers of the longest drill core characterised. Feasible physical separation methods selected were Knelson concentrator, LIMS and HIMS. In the Knelson concentrator, 56 wt.% tungsten was recovered in the 23 wt.% of concentrate produced. In magnetic separation, 82 wt.% tungsten and 91 wt.% copper was recovered in the HIMS magnetic fraction, while sulphur was only 1 wt.% in the HIMS non-magnetic fraction. This study is fundamental in the development of effective methods for separating minerals and extracting metals of interest from these historical tailings so that an inert and environmentally safe residue is left behind.
Niobium is an important element that has been applied in many cutting-edge applications. Currently, its beneficiation is through froth flotation, although the fines are lost as slimes, comprising up to 15 % of the total niobium content. Recent work, described by Mark Lepage, of McGill University, Canada, has shown that gravity and magnetic separation are effective for processing niobium-bearing minerals. However, studies using these techniques in the ultrafine range (below 10 µm) are still limited. The work investigated the feasibility of using enhanced-gravity and magnetic separation techniques to extract niobium-bearing minerals (columbite and pyrochlore) from industrial slimes containing gangue such as dolomite. The results from magnetic separation were then related to the magnetic properties of each individual mineral.
A recent trend in wet processing of incinerator bottom ash is the development of a fine ash treatment module for size fractions - 2 mm. Options for the extraction of the fine size fractions in both wet or conventional dry process plants were presented by Jonas Boehnke, of Sepro Mineral Systems Corp, Canada, and the economical environmental and health and safety aspects of extracting and processing the fines were discussed. Ash samples are exceptionally heterogenous in their composition and this particularly applies for precious metals. Particle shapes also play a significant role in the selection of suitable recycling technology. Comprehensive testwork at laboratory scale has identified a compatible gravimetric separation route and it is explained why only a combination of multiple devices satisfies the diverse requirements to produce a marketable heavy non-ferrous metal product. Experience from laboratory and prototype operations has shaped the process and led to the successful launch of a commercial scale plant at Brantner Oesterreich GmbH‘s operation in Austria.
With the ever increasing need to conserve water, efficient dewatering is a priority. As discussed by Christian Ihle, of Universidad de Chile, this is particularly challenging for dewatering slurries with a large amount of fines. Chilean mining highlights the global trend of decreasing ore grades, with the existence of operations which are located in desert areas, thus having limited access to water close to operations. Christian described numerical simulations of the concurrent impact of Boycott effect and differential heating at the downward facing wall of a lamellar settler operating both at batch and continuous mode. Both types of operation reveal a general enhancement of performance with the amount of heating in the particle accumulation at the bottom for a specific particle size range. However, a trade-off between enhanced settling (and thus system capacity) and re-suspension near the top is observed. Opportunities for scaling up to industrial infrastructure were discussed.
The trend in tailings dewatering has always moved towards a drier product, recovering more water for economic, environmental and social pressure, which are especially important in recent times with tailings disasters being brought to the public eye. This has been achieved through the adoption or application of technology, starting with the implementation of thickeners and the subsequent optimisation which lead to the development of Paste Thickening.
Ten years ago this was very much the "buzz" word of the tailings industry. It seemed that every site, both greenfield and brown, was considering Paste Thickening of their tailings, and vendor pictures of thick toothpaste like tailings removed from the bottom of a pilot plant were a common at conferences. Is this still the case today? Not according to Paul Ainsworth, of Paterson & Cooke Ltd, UK, who said that many mines while operating paste plant above the typical underflows achievable by other thickening technologies do not produce the envisaged thick, stacking tailings that was so prevalent in their advertisements. This is not to say all paste installations failed, on the contrary some are very successful, but not all.
Ten years on while current investigations still consider paste it appears to have taken more of a back seat to the next stage of dewatering, filtration and the tailings storage that comes with it, Dry Stack.  As filtration and dry stack tailings is becoming the new “buzz” phrase of the tailings industry and is under consideration for many projects Paul asked how can it learn from Paste Thickening. He highlighted some of the areas that need to be improved, expanded and better integrated so that dry stack tailings is more successful, avoids the pitfalls and prevent projects from succumbing to the failings that some Paste Thickening operations experienced.

Paul Ainsworth (right) with Kathryn Hadler and Tom Naylor
Paul's topical paper ended the many fine presentations over the two days, and after Amanda thanked everyone, particularly the sponsors, for their contributions, I gave an introductory talk, supplemented by a traditional Cornish cream tea, on the geology and mining of Cornwall, a prelude to our short visit to the site of the 19th century Wheal Peevor workings.

By all accounts this has been a fine conference with some very high quality work presented. Once again many thanks to our sponsors, and to the chairmen who kept the conference running on time despite the many short presentations. I would also like to thank the staff of the National Maritime Museum, particularly the tireless Dayna Pope, for making our debut at this magnificent facility so enjoyable. I would very much appreciate feedback from all who attended.

A full set of papers is available from MEI as well as a full set of conference photos.

Physical Separation '21 will be held in Falmouth again in June 2021. Full details will soon be available on MEI Online. Hope to see you there!

Twitter @barrywills


  1. Thanks you Barry for such an exhaustive report.
    The work on application of microwave is very encouraging about which we have been reading over the last 30 years but not much came out; the papers presented covered studies on minerals and application at industrial scale--augurs very well and gives confidence for others to quickly adopt in industry.
    All the other papers also seem to "hit the nails at the right places" --a paradigm shift --explanation of need of industry and solutions which are implementable leading to better practices.
    Seems theory and practice are coming together--augurs well to mineral industry.

  2. Thanks Barry and all MEI staff for such a great event!
    It was my first MEI conference, and hopefully the first of many to come!
    The involvement of several mining companies created a great opportunity to discuss the importance of our research work!
    It was amazing to see recent technologies being discussed, like microwave and electric pulse fragmentation, simultaneously with “old” technics as screening and sorting - concepts that are being recovered and improved with new technologies! Due to the growing environmental concerns, physical separation seems to be the most suitable solution for a low impact mining, and the Physical Separation’19 conference was proof of this.

    1. Thanks Rui. It was great to meet you for the first time in Falmouth, and we greatly appreciated your very active involvement. Look forward to seeing you at future events


If you have difficulty posting a comment, please email the comment to and I will submit on your behalf