Monday, 5 May 2014

Integration is the key to the future of comminution

Comminution '14 ended with an excellent panel discussion on the future of comminution, admirably chaired by MEI consultant Dr. Aubrey Mainza, of the University of Cape Town. The four panelists were Dr. Rob Morrison and Prof. Tim Napier-Munn, of Australia's JKMRC, Prof. Marcelo Tavares, of University of Rio de Janeiro, Brazil and Prof. Wolfgang Peukert of University of Erlangen-Nuremberg, Germany.

Aubrey Mainza, Rob Morrison, Marcelo Tavares, Wolfgang Peukert and Tim Napier-Munn
Following are highlights from this discussion:

Prof. Peukert began by saying that much could be gained if the science of comminution and industry could come closer together. It is apparent that modelling now gives us a greater understanding of what is going on in a milling circuit, and there is a lot to be gained from detailed modelling. The models, however, must be checked by reality to give us a reliable toolbox to assess what is happening, particularly with complex multi-phase particles which can be characterised to assess liberation, the balance between strength of grain boundaries and strength of grains. On a larger level classification and mill throughput are critical, as they affect energy efficiency and overgrinding and we need a better understanding of the transport properties and an ability to remove particles which are properly stressed. We do not yet know how the particles are stressed and we have to accept that we do not as yet have ideal models for process simulation. He says that the Universities are prepared to undertake work on these problems, and if industry comes in as well we can solve a lot.

Prof. Tavares said that it was important to compare what we saw two years ago at Comminution '12, particularly in crushing and dry comminution, and the potential now for grinding finer using compression crushers, which is important with the need to conserve water in our industry.  We have seen this year different levels of comminution modelling living together, from the very basic Bond equation, which has its place in the minerals field, to the population balance models of the 1960-80s, to the advanced models of today. He also stressed that it would be good to see at Comminution '16 some strong case studies with industry results to demonstrate the effectiveness of these modern methods, so it is important now for the researchers to strengthen their relationships with industry. He stressed again that we must increase our understanding of machine and material contributions, particularly fracture response and its contribution to liberation.

In Prof. Napier-Munn's opinion, Comminution '14 had been an exceptional conference. He had "been blown away" by the vigour and enthusiasm of the smart people who are doing great work in a very mature field, and it is extraordinary to see new ideas still coming along.  He said that in terms of the future of comminution "we really have to get rid of tumbling mills". The energy crunch will soon come and it will be very bad for our industry and we must put the next few years to work with the people who attended. It is obvious that there are many new ideas out there. New crusher developments, HPGRs, fine grinding devices and circuits offer new approaches and some hope for the future.

As with the others, he expressed his disappointment that the heavyweights of the mining industry did not attend the conference, with the exception of Anglo American, who were present in force, and unless the operators, who are the users, drive these new ideas in a constructive and intelligent way, then we will not make progress as fast as he thinks we need to.  He enthused about the modelling, which he feels has now given us a mature set of tools. DEM, CFD, SPH and their combinations, are now ready to solve problems, which was not true 20 years ago. However they are only being used to solve problems in a few patchy areas, and the challenge to change the comminution scene is to make much better use of these models with some real integration between the modellers and the people with the problems, and this must be driven by the companies.

Jeremy Mann, of Anglo American, South Africa,  remarked that we had fantastic tools available now, but we should look more towards understanding selective liberation rather than just breakage, and to use the models in process control systems, which means that we need better understanding of process mineralogy and the variability of our ores.  Rob Morrison agreed with Jeremy, and noted that the few people at the conference who had talked about liberation had all been interested in the liberation of the valuable components, but the real challenge is to liberate and discard the major proportion of the ore, the gangue, at a coarse size. He congratulated Chris Greet for presenting a flotation paper at a comminution conference, but then said that if people followed his strategy most mines would go broke very quickly! He was disappointed that no one had mentioned at the conference that in rougher-scavenger grinding we are actually trying to liberate the gangue, we cannot afford to liberate the valuable minerals, and the most difficult decision is how far to grind, and this is largely an economic one.

Dr. Greet responded to this by agreeing with Rob, in that we only try to attain 60-70% liberation, dependent on the mineralogy, but we aim for much greater liberation in the regrind circuit. He made the point that comminution is not just crushing and primary grinding but regrinding as well, so basically both he and Rob are right!

Adrian Hinde of Mintek also highlighted again the need for integration between the researchers and industry, and also argued the case for more integration between the mining engineers, dealing with blasting, which is essentially the first stage in comminution, and the mineral processors.

I think the key message here is that there is a need for greater integration between academia, research institutes, suppliers, all of whom were well represented at Comminution '14, and industry, which was not. Hopefully the operators will be out in more force at Comminution '16. There is much to be gained by their presence, preferably to present case studies on operating plants, validating the work of the researchers, or just to network. As Clifford Mutevhe, of Anglo American, Zimbabwe observed "(Comminution '14) not only enhanced my knowledge but also helped me identify solutions to the milling circuit I manage. Some of the discussions we had we have carried forward as part of the business process improvements. I also managed to identify with a huge pool of knowledgeable people from across the world. I could not imagine communicating daily and interacting with people from RSA, USA, France, Iran, China, Australia and various other countries."



  1. Thanks for this very interesting summary Barry. I agree fully that more integration is needed between the researchers and industry, not only in comminution but in all areas of mineral processing.

    I could not attend Comminution '14 but your superb report brought it to life, and I obviously missed a great event.

    Malcolm Powell's GCC initiative is splendid, but I noted from your report that all the collaborators are from academic establishments. Would this not be a great opportunity to get the big mining companies on board- Anglo, Glencore, Rio Tinto, First Quantum (who I see was a conference sponsor), etc? This could pave the way forward to real integration between research and industry.
    D. Robinson, Qld, Australia

  2. Good to know some one has taken a step to understand the importance of energy required for grinding, and bridging the gap between R&D --Industry. Congratulations to all team members of this discussion.

  3. A lot of work was done on modelling liberation previously

    The methodology for solving this problem is based on information theory (probability theory). I explain many of these advanced concepts in a course I give 'Simulation of Mineral Processing Plants'.

    In addition I have developed I have developed a simulation system MMVisioSim, which is based on multimineral particles.

    The software, and the courses are all aimed at better understanding of mineral processing plants,

    In particular the software is highly extensible meaning different models are available for the same units. For example , a researcher may wish to model a unit in great depth, and an operator may wish to consider only the operational parameters.

    The software I have developed provides linkage of the numerous steps required to improve plant performance. The particular niche of the software is inference capability which allows to estimate detailed multimineral particle information using advanced mathematical methods (i.e. Hidden Markov Models/patent pending).

    One of the problems is that many academics recognise the importance of liberation but don't have the necessary mathematical knowledge.

    This is why the course I provide is so good, as I get the whole spectrum of participants inclusive of operators and lecturers.

    The software is likely to gain better profile through an AMIRA project; but anyone interested in progressing independently should contact me directly. Indeed many good ideas emerged from a particular gifted PhD student Jon Keith who later went on to model DNA sequences.

    As more researchers get involved there will be greater opportunity to publish, so hopefully we will be able to publish these new ideas in a few years time at the various Minerals Engineering Conferences.

  4. responding to anonymous; D Robinson?
    "The GCC is sponsored by industry and industry provides the facility, the actual plant and its survey and operational data to develop the work - Anglo are major sponsors of the work being done by Malcolm and his expert group!"
    A general comment regarding focus - financial motivation is the focus for improvement at the bottom line level for metals recovery/gangue and values liberation within circuits - rejection of gangue is crucial at each stage of the flow sheet, maximised, hopefully before the energy input intensifies, ore upgrading in ROM; thereafter its the balance of optimising mainstream metals recovery/gangue rejection and then optimising side stream grade and recovery. Its a question of balancing the flow sheet chain of gangue rejection and product grade to be the best overall financial solution; costs/revenue. Also worth remembering flotation products then go onto higher energy intensity processes; e.g. smelting; to finished metal production

    1. Thanks for your clarification on the GCC, Chris. It was good to see Anglo so well represented at the conference, with 8 delegates from RSA, 2 from Brazil and 1 from Zimbabwe. Hopefully there will be more industry involvement at Comminution '16

  5. Malcolm Powell6 May 2014 at 13:42

    Thank you for the favourable comments on our GCC initiative. Pooling our limited research resources is a key driver to this group. As Chris points out, Anglo are our major sponsor - with funding that aims to integrate fundamental research and applied process improvements - as published at the SAG11 and Mill Operators 2012 conferences. We are also working with a number of major companies via the AMIRA P9P project, plus equipment suppliers. I strongly agree that integrating R&D across the process chain with the associated companies (equipment, engineering, process consulting, mining and control) should be a major driver to improve the status of the mining Industry - as outlined in my opening address.
    I hope we all left the conference re-inspired for the challenge!
    Malcolm Powell

  6. There is also discussion on this topic on the MEI Minerals Engineers group on LinkedIn

    1. With a particularly insightful view from Jarrod Hart, of Imerys, UK

    2. And there is also discussion on this at the CEEC group on LinkedIn

  7. From my experience with minerals processing conferences, the MEI's Comminution is really at the top level of quality presentations covering academic research, industry experiences and future trends that provide essential course corrections to the minerals industry rarely seen by the corporate executives and investors. Many thanks to the Wills family, the sponsors and all the active participants.

    The panel discussion was a good wrap-up to the conference and indicated the very different views and perspectives present. I think it was Aubrey that matter-of-factly stated that after decades of research and centuries of industrial use, we still don't have the modelling right for comminution. A sad truth when we want to think about efficiency standards and flowsheet benchmarking, primarily to reduce energy consumption but secondly to the benefit of the industry, the plant operators.

    That tumbling mills need to be replaced as mentioned by Tim, is music to a vertical stirred mill supplier's ears, but the industry is not known for its speed in adopting new, superior technology on a global scale. Tumbling mills will be here hopefully long enough for some of the simulation and mathematical models to improve their estimation accuracy for traditional ball mill performance, which equipment suppliers like Eirich would be happy to adopt and apply to their mills performance estimations, if the validity of these models has been proven for the mill type in question and once our customers have confidence in these models. Remember that the equipment supplier is liable for performance deficits, not the academic research institutes.

    I agree with Professor Peukert that science and industry need to come together, and all comments supporting the integration-is-key topic. We need to remember that all of our efforts, both those in the industry and those on the academic side, are driven to provide an economic benefit. Whether less gangue is being processed, or less energy consumed, suppliers need to sell and operators need to produce at a profit. With that reality in focus, the academic efforts and energy-saving initiatives will be welcomed partners of industrial operators and equipment suppliers.

    I, too, hope to see much greater industry participation during Comminution 2016. All the best till then.

  8. Although I did not attend the conference, my views are as follows. Tumbling mills will never be replaced, they are masterpieces of engineering and do a lot of work for us. Producing and trading commodities will become more of an issue in future. Countries will produce what they can within their own borders, and funding will come from Governments and tax payers. The ability to fund ever larger projects is going to be the main issue. Smaller operations/companies may evolve, but not at the same standards as their 1 companies. I think Ian Runge is on the mark, and in particular slide 7. see This is an excellent precursor to the future.


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