Thursday, 11 April 2013

Froth Flotation. Why is it still so intensively researched?

Try explaining nuclear fusion or a car's SatNav to a non-scientist. It's not easy. But explaining the principle of froth flotation is not too difficult- you add chemicals to a mixture of ground ore particles and water, which makes the valuable particles 'greasy' and thus water repellent. You then blow air through the mixture, and the greasy mineral particles stick to the air bubbles and can be skimmed off and separated from the waste.

Simple! So why does this apparently fairly simple process, which was patented over a century ago, attract some of the finest scientific minds in the minerals industry and contribute to a major portion of the peer-reviewed papers published each year in mineral processing journals?

This posting is prompted in part by that of 27th March, and also by an email from Dr. Stephen Grano, of the Universityof Adelaide, Australia, one of my most respected Minerals Engineering reviewers.

Stephen Grano Froth Flotation
Stephen writes: I think a lot of scientists dabble with flotation in their career because it has elements of hydrodynamics, surface chemistry, plant design, froths, etc for a three phase system with mobile gas/liquid interface. So it lends itself to research. But I sometimes wonder where we have got with all the research. The fact is that nowadays a plant metallurgist should be getting >90% recovery and should consider their own position if they are not, with all the tools (QEM-SCAN etc) that are common place. Thirty years ago that was not the case. Thirty years ago people suspected that pyrite in a copper concentrate was activated by Cu ions - now they have proof. But what to do about a plan? I believe that the next step changes will come from new sensors which will be able to measure liberation on-line and surface chemistry on-line. If I were a company, I would not be putting my money into standard surface chemistry research any more. They need sensors in the plant. They know the mechanisms now. The trouble with current sensors like x-ray probes is that it is only a bulk assay and not a measure of liberation. So we are at 90% recovery but can we maintain that for all ores and at lowest possible cost? That means we need to maximise throughput for recovery which means floating coarse composite particles of low liberation. That is another area for research. Do we need more review articles on copper activation and oxidation, and water effects? Do we need more papers trying to increase fine particle recovery by hydrodynamics as it is possible to recover these particles given enough time? We know a lot of that already. One of these days, either next year or in 10 years time or longer, someone will have to say "this is the same as what was published in 1985 or whatever". What I am concerned about is papers that purport to be different but in reality are very derivative or that are pseudo fundamental and justified as being fundamental but in fact don't shed any new light on the matter. I am afraid to say that there are so many now doing flotation that the research has been somewhat commoditised in my view. It may be that more rigorous reviews are required. I get the impression that some researchers think that if it was not in the last 5 years it has not been done. Yet the fundamental questions remain unsolved and continue to be unsolved.

These are provocative words from Stephen, which I hope will promote some serious debate both here and on LinkedIn. Is part of the answer to my eponymous question, as Stephen implies, regurgitation - are journals publishing work similar to that carried out in the past, but now lost in the mists of time? Are we to some extent going round in circles?

I have a feeling that this might be partly true, and is evidenced by the plethora of flotation papers churned out each year at international meetings.

But there is more to it than this. I have argued before about the importance of flotation, indeed that it has my vote for being the most important technological development of the 20th century (6th June 2011). Originally developed to concentrate sulphide minerals, it is now ubiquitous in the minerals industry, treating oxidised ores, oxides, non-metallics and coal. Miners have always 'cherry picked', treating the richest and easiest ores available, such that today the available ores are often low grade and very difficult to treat, and many modern industries demand metals which were virtually unheard of 20 years ago, such as indium, lithium, germanium and neodymium. Intensive research is therefore needed to keep pace with this increasing pressure on the minerals industry.

Today's civilisation is dependent on a plentiful supply of the so- called 'base metals' such as copper, zinc, lead and nickel. Without prior concentration of the valuable minerals in the ore, extraction of the metals from these minerals would be hopelessly uneconomic, effectively making these base metals 'precious metals'. The wealth of information on the flotation of these ores is vast, but even so intensive research on sulphide mineral flotation continues, in the quest for more effective and environmentally friendly reagents, improved circuits and machines, and better ways to control and optimise the process. Base metal mines treat huge tonnages and very small improvements in recovery can lead to huge gains.

In November the world's leading flotation scientists and engineers will meet in Cape Town for Flotation '13, MEI's largest conference. Papers will be presented on the state of the art in flotation technology and the meeting will provide the opportunity for young researchers to network with the major players in this field. The deadline for abstract submission is the end of June and I look forward to receiving papers on cutting edge technology, hoping also that authors make use of the wealth of information stored in the archives over the past 100 years.


  1. I have to agree with many of the aspects Stephen Grano listed.

    Why for research?

    1) The high economic value of improving performance. The fact remains that there is tremendous value achievable from small performance improvement. The 90% recovery Grano mentions as achievable remains a distant beacon for many operations.

    2) Intellectually interesting: Although superficially a simple process, the many interactive aspects of the process and the potential to work at scales from the micro to macro continue to make it challenging and interesting in application and for improvement.

    3) Failure to recognise / consider past work in solving present problems. This is seen to be the case even when the present problem has been solved previously somewhere else.

    Admittedly some work can be more difficult to access. However, I find people in general interested in discussing their work and the effort to look and understand should be at least be made as a starting point.

    The fact that many researchers publish the same work in various forms, various places, etc. does not assist in following developments. This particularly as much of the work is as he says (simply) derivative.

    4) Failure to define research problem before initiation of work. The goals of much research seem poorly defined / not easily related to actual use of the process.
    Robert Seitz,USA

  2. I would also partly echo with Stephan's comments. That new people in the organization do not pay much attention to classic work in the literature. It could partly be due to the fact that not many true mineral engineers are researching the flotation. There are plethora of students from different fields than mineral processing joined the industry in search for job, and therefore do not understand the concepts completely. And whatever is researched is published. So you see plethora of papers on flotation for the similar concepts some 30 years ago.

    I believe in some way there is a communication gap between the researchers trying to understand the hydrodynamics effects separate from the surface chemistry effects. The industry perspective is mostly concentrated in developing the mechanical improvement in the float cell to get major share in the market. On the other hand, academic leaders focused mainly on the surface chemistry effects. They do come closer only at few occasions, such as FLOTATION conference. So, you see the homogenized hypothesis is looked differently between industry and academic leaders, and therefore the gap in the understanding.

    As Wills mentioned, the ore is now getting more of low grade and poorly liberated. We have to look for the pre-concentration before ore could be send to the plant for further processing. There has been some progress in terms of on-line analyzers such laser induced breakdown spectroscopy, ore sorting, geometallurgy concepts etc. Hopefully, we will see more developments in this area.

    With the technological advancement, the devices are getting much smaller and faster in terms of processing. There is good interest for new metal properties that is available in the rare earth minerals, which should be studied and explored more.

    Vishal Gupta, USA

  3. I think it is interesting to compare the level of academic interest in flotation with the level of interest in fine gravity separation. Both areas are highly significant to coal and minerals processing, and both are relatively mature areas. I am sure that academic interest in flotation exceeds that of gravity separation by 1 to 2 orders of magnitude. Flotation is academically the richer due to the importance of chemistry, physical chemistry, and hydrodynamics, and lends well to research training. I think that in an area as large as flotation there will be a good bit of mediocre work, but that will be true in many fields. In my view there are many important challenges that remain.

    The future of mining will be increasingly focussed on the recovery and concentration of fine particles (due to falling grades), and in my opinion there are many challenges that remain in both flotation and gravity separation. Gravity separation needs to perform better across a wider size range, and to perform well at much finer sizes. Similarly, I think flotation needs to perform better across a wider size range, but must also cover coarser sizes. Desliming is critical to both, and can present very significant problems. Increased kinetics and throughput are key issues, which must be investigated in the context of achieving target grades and recovery. There is plenty more to be done. What we need to do is to ask the right questions and hence identify the right problems for solving, and develop more innovative approaches.

    Kevin Galvin, University of Newcastle, Australia

  4. Morning Barry from the wilds of south-eastern British Columbia:

    Stephen is well known to the minerals research community, not least through his prior association with the Wark Centre at the University of South Australia.

    Stephen makes valid points concerning the need for technology for on-line measurement of liberation. Some years ago there was a conceptual approach at the JK (Geoff Lymann?) using a wide spectrum analyzer (IR through UV) using novel microscopic techniques to provide rapid liberation analysis of core. I do not believe that this idea went further than the concept stage, but it potentially would provide an approach to Stephen’s query.

    Stephen's comment that mineral process engineers in plants should achieve 90% recovery is perhaps controversial - depends on the ore - see our paper comparing the flotation responses of Polaris, Sullivan and Red Dog in galena and sphalerite flotation with respect to pyrite and pyrrhotite separation. Where liberation was not an issue (Polaris) exceptionally high mineral separations were achieved; at Sullivan and Red Dog, imperfect mineral liberation constrained and continues to constrain the mineral separations achieved. With favourable pulp chemistry, the separations of free minerals was similar.

    The efficiency of the flotation process as a function of particle size is well known as is the relationship between residence time and flotation both as a function of particle size and mineral hydrophobicity i.e. as residence time is increased to recovery fine, free particles, mineral separations become limited by the undesirable deportment of partially hydrophobic minerals; this is problem, particularly for differential flotation – Cu/Pb/Zn/Fe, Pb/Zn/Fe, Cu/Fe.

    The need for constructive analysis of problems based on a strong viewpoint to prompt critical thinking is of course no bad thing.

    Stephen’s point about solved problems continuing to be investigated is of course valid. It seems to me that this is one of the roles of critical reviewers – to determine if research papers truly contribute to our knowledge base. It would be nice to be able to intercept planned research programs at the concept stage to help direct research energy, time and money to problems of both fundamental and practical importance.

    In reflection, it would be interesting to hear Noel Finkelstein’s comments to Stephen’s note. Noel’s review papers remain key reading for any researcher.

    Kind regards
    Mike Fairweather
    MJ Fairweather & assoc Inc>Rossland, B.C.

    1. You are dead right, Mike, regarding the 90% recovery target. There are other factors besides the efficacy of flotation, mainly the mineralogy as you point out, and, most importantly, the efficiency of the comminution process. It is no coincidence that MEIs three major conferences are flotation, comminution and process mineralogy.

      The point I was making about research into base metal flotation is that the tonnages treated are so vast that even very small improvements in recoveries can lead to enormous cost savings, justifying continued research efforts.

  5. This outcome of many papers that only describe small increments of knowledge and learning tends to be encouraged by the academic funding model.

    In Australia, Academics are funded and rewarded for publishing papers. So this would tend to lead to every small advance being published in a separate paper, as this is what the funding model encourages. Also I think this type of funding model is common in the universities of the western world, and may also be common worldwide.

    But I do like to read the individual steps, as you get more detail on the process being investigated, than sometime occurs in summary papers.

    Michael Young, Xstrata Technology, Australia

    1. Michael, it appears the limits on academics you mention are prevelant across the west, and likely the rest. Given how much of the work comes from a combination of academics and sales people, the state of technical literature should be no mystery.

      Achieving a sense of integration across the whole of flotation research is part of the challenge with this research.

      Aspects of:
      - the physical environment - equipment / hydrodynamics
      - the chemical environment - solution and surface chemistry
      - the ore - valuable mineral(s), gangue mineral(s)

      are often dealt with independently and with experimental designs that do not lend themselves to understanding the interaciton of variables.

      Many researchers appear to ignore this, to fail to clarify how their increment should be integrated into a whole. My hypothesis is that they are unaware and part of this is the complexity / nature of the subject.

      As Michael mentions, it's good to read the individual steps of learning to appreciate the detail; and I'm sure we are all working to integrate this knowledge into a vision of the whole. I hope we are not like the blind men explaining (or exploring) their view of the elephant!

      Robert Seitz, USA

  6. All this discussion (and the decades of research) is testament to the importance of the Flotation process, and its robust and reliable nature. Despite its considered maturity, it is still the preferred process and as long as there are aspects we don’t understand and the potential for improvement, there will be ongoing research. As deposits become more complex, operating environments more stringent and technology offers us better options, I suspect the amount of research will grow not diminish!

    Best results are obtained where we all, practitioners and academics, work together and the ongoing support and increasing interest in the MEI Flotation Conferences provides the perfect platform.
    So I look forward to continuing this discussion in person in November!
    Dee Bradshaw, JKMRC

  7. I agree with Stephen when it comes to the fact that older pieces of research (especially those older than 30 years) are not as known as they should be. I think there are several issues involved:

    1. Part of the problem, as was already mentioned in this thread, is the availability. In the old days, conference proceedings were considered as prestigious as journal papers. Perhaps this is a reflection on the standards of peer review as conferences get larger. Nonetheless, many of these old proceedings are not easily available online. One can still get them through library services, but there is no instant download option available.

    2. The writing style of academic papers has changed dramatically over the last few decades. Reading some of the older articles, one gets the distinct feeling that readability was not high on the author's priority list.

    This is not a problem if you are an experienced academic with a well built up knowledge base, which allows you to spot the gist of the article without getting bogged down by detail and language. But, this is a serious deterrent for younger researchers and students. In my experience, the vast majority of students seriously struggle with comprehending older publications, while not having similar problems with newer ones. It is not then surprising that these up-coming researchers base the majority of their knowledge on more recent work, resulting in a progressively increasing lack of citations of older work. Then a new crop of researchers comes along and only look at recent work which only cites more recent work, thus perpetuating this collective research "amnesia".

    The solution to this is that it is up to the more senior researchers to invest more time into junior staff, to make sure that they are aware of the work that took place decades earlier and of its significance. Perhaps what we need is an annual special journal issue, where people nominate and discuss an old piece of work which they conciser to be important yet largely overlooked (not their own!), thus reviving/highlighting some golden oldies. What do you think Barry, would that work?


    Liza Forbes, CSIRO, Australia

    1. Your idea of a special 'golden oldies' edition is an excellent one Liza. I will give this some thought.

    2. Apropos of reviving old words of wisdom....

      (1) I agree that conference papers are well worth reading. For a start, some conference papers never get 'converted' into journal articles.
      Also, as Liza alluded to, some conference papers — particularly those of the pre-internet age, and especially those that recorded subsequent discussion — deserve to be considered as being at least as prestigious as journal articles.
      ↳Anecdote A

      Liza is correct to say that availability can be a deterrent, but at research organisations with good librarians and library systems it needn't be: most often it's just a matter of entering the bibliographic details into a form on the library's website and waiting for the PDF to appear in your email inbox a few days later!
      ↳Anecdote B

      (2) I agree that the writing style has changed. I don't think I would say that I find old papers harder to read personally. If you go back far enough the two things I would say is that styles were more conversational, in general, and also more idiosyncratic. Perhaps having less of a standard template to follow in the past might make things more difficult for (modern) readers.

      (3) What to trust? There will be _some_ old publications whose results were based on technology or theories now known to be unacceptably inaccurate by modern standards. Researchers may be 'assuming' (hoping? wishing?) that any valuable contributions from past ages have already been incorporated into whatever new publications they read. It saves them time, and also means that they don't have to make a reassessment of the old article themselves.

      What to do about it?

      Highlighting 'classic articles' is a valid idea. However it might end up spotlighting just a tiny proportion of the old works — perhaps even those that still tend to be cited anyway!

      At a broader philosophical level I suggest that more emphasis be placed on the value of going back and reading the original publications. Both to credit the originators, and to gain more knowledge oneself. Practice of citing 'classic' works without ever even seeing them is supposed to be known to be wrong, but it still happens.
      It is not just 'classics' that deserve to be read. If you are relying on something that is justified with a citation, go and read the cited paper. If the idea/data didn't originate there, then go and look at the paper that they cited, and so on back to the source.
      Of course, this takes time, and probably the payoff is too far down the track to benefit PhD candidates who are pushed to publish as many papers as possible as quickly as possible.
      Therefore one of the biggest incentives to do it would be if reviewers look for it and ask for it. "It" being not just the meaningless addition of more citations without context, but rather citation with proper critique of the work cited.


    3. /cont
      I once (this decade) heard advice from a senior researcher that if you have a manuscript that you're 'not entirely confident in', then send it to a journal, not a conference. The logic being that if your lack of complete confidence was well-founded, and there were indeed problems with the manuscript, then in the first case only the editor and (perhaps) a couple of reviewers would see it; but in the latter case everyone would see it.
      That advice makes some implicit assumptions about the current reviewing processes/standards in each case too.

      I once knew a recently hired research fellow (i.e. PhD graduate) who told one of the senior research fellows that he hadn't read a certain paper because it wasn't available online ...although it was available in print in the library ...which was less than 200 metres away from his office. The senior research fellow apparently told him exactly what he thought of that attitude.

      Even if an old (or new) publication used a faulty analysis, if the observed data themselves are presented and are trustworthy, then they can be reanalyzed with state-of-the-art tools. Alternatively, if the measurement technology used is (now) known to have been biased in a specific way, even inaccurate data could become informative.

      David Verrelli, Macquarie University, Australia

  8. Flotation research is very important as we know. Millions of ore need to be concentrated to be able to prepare the feed for smelting. As such eliminating all the gangue. Flotation is the separation of mineral using hydrodynamics and physical chemical properties of minerals.

    We have a dynamic simulation approach to understand flotation. You can take a look of Dynafloat in the App Store.

    These simple models integrate many of the most common subprocesses involved and their interactions.
    Check out this video on YouTube: Dynafloat video . Ipad version

    Osvaldo Bascur, Chile

  9. Grinding is the science for flotation to work properly.

    First we need to liberate the mineral species from the ore.
    As such, then flotation picks the right species to be carried by the bubbles.
    But, nothing is perfect in nature. As such, some middlings will attach to bubbles and some contamination will exist reducing the grade of the concentrate.

    In addition, the fine gangue particles will follow the water attached to the bubbles and entrain these non wanted particles reducing even more the final concentrate grade.

    As such, the froth high is critical. The bubbles brake and produce a strean of watee going down, this stream of water clean the non wanted particles and put them back into the tails. The tails are descarted into waste.

    So, to see how operate your typical gringing circuits we have dynamic virtual simulatirs to tune the grinding processes and particle classification hydrocyclones to get the right size of liberated particles for flotation.

    If you want to see how this work in industrial practice you might want to try the grinding game. Be careful not to use to much water or not enough oherwise the mill will choke. This is the same principle if you eat large pieces of food. You always need some water to clear your throat.

    Check out this video on YouTube: Dynamill

    For ipad version found in apple store:
    Osvaldo Bascur, Chile

  10. Explaining the principles is not too difficult indeed.
    The issue is that the technology has limitations (too small or too coarse particles do not float efficiently), which are beyond the pure phisico-chemical area and more related, for example, to the hydrodynamics of the process. Technology to measure variables related to cell design (turbulent energy dissipation rate, RTD of gas, liquid and solid/size-class, etc.) and its influence into flotation efficiency is still a challenge. Economics behind a "marginal" improvement (recovery or energy efficiency) are enormous given the extent of the technology utilization (only in Copper, 2 billion tons of copper ore are floated yearly).

    Adrian Villanueva, BASF, Germany

  11. Barry's original question can easily be asked of much of the reseach underway across most fields of investigation. I hesitate to say all since only look at some fraction in mineral processing and some elsewhere.

    That leads us to asking the goal of funding academic research and the split goals of developing new investigators vs. new knowledge. What's the right balance between these goals?
    Robert Seitz (USA)

  12. Hi Barry,
    I readily agree that flotation is a key separation process in the mineral processing discipline, and that it is complex and interactive. Much more still remains to be learnt before we develop a more predictive capability on this unit operation. However it would be remiss to exclude the very primary stage that the comminution circuit sets for this flotation process. As many researchers have ably shown, the size distribution presented to flotation has much to do with the potential flotation performance. That said, I will single out the use of mixed collectors as one of several fields of good opportunity in the advancement of flotation. As Professor Bradshaw has already noted on this post, our global collaboration in sharing our findings as we progress on the path of better understanding of flotation is important. The MEI Flotation conferences in Cape Town are a good venue to do this.
    Dr. Norman Lotter, Xstrata Process Support, Canada

  13. Flotation sensors. I couldn't agree more Stephen,
    It would be nice to see a research program focused on flotation sensors, ultimately to create liberation and surface chemistry sensors; but also to develop low cost, on-line sensors capable to provide basic information such as concentrate flow meters on every cell or information regarding pulp hydrodynamics and froth behaviour. From experience I can say this kind of research programs require flexible minds, willing and capable of understanding both sides: instrumentation and froth flotation. This is a really interesting and worthy cause; count me in if ever such a program is born.
    Carlos Vanegas. Australia

  14. This is super interesting. Early in my carreer as a Mineral Processing Engineer we instrumented heavily a Flotation Cell to understand its behavior.

    This research lead to the development of an online model for flotation. As such, we can estimate with on line data the liberation and the different kinetics of the mineral species. At the same time the online model can be used to predict the content of metals on line.

    For addtional information see The Flotation of Sulphides Minerals by Eric Forsberg. On line Flotation Model. Bascur and Herbst. University of Utah.

    Now, we have simplified the dynamic model. With the new advances in communication and real time information systems this is becoming a reality. Any interest in it?
    Osvaldo Bascur, Chile


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