Process Mineralogy '14 Conference Diary

Process Mineralogy '14 was held at the Vineyard Hotel, Cape Town from November 17-19, 2014. The 3rd in MEI's series of process mineralogy conferences, it was also the smallest, reflecting the current depressed state of mineral commodities, particularly gold and PGMs, but we thank the companies who provided generous corporate support:

Monday November 17th

Jon opened the conference this morning and welcomed the 74 delegates representing 15 countries. Numbers are down in comparison with Process Mineralogy '12, but most of the companies and institutions that were present two years ago are represented, albeit with fewer delegates from each institution- a sign of the hard times.

MEI Consultant Dr. Megan Becker, of the University of Cape Town, then summarised what we have in store over the next few days, indicating that using mineralogy for prediction is a major theme of the conference.

Megan is back in Cape Town after spending the last 2 months at Luleå University of Technology (LTU) in Sweden as a visiting scholar on the Emerald International Georesources Engineering Erasmus Mundus Masters Programme. The International Masters programme is run through 4 European Universities: University of Liege (Belgium), University of Lorraine (France), Luleå University of Technology (Sweden) and Technical University of BergakademieFreiberg (Germany) entailing the students spending time at each of these institutions. Hosted by Prof. Pertti Lamberg, Prof. of Geometallurgy at LTU, her visit entailed teaching and interacting with the international class of students who are currently based in Luleå studying geometallurgy and mineral processing.

Megan Becker (centre seated) and Pertti Lamberg (2nd left seated)
at LTU with international students

And it was Luleå University's Pertti Lamberg who got the conference off to a fine start with his opening keynote lecture, discussing the use of automated mineralogy for modeling and simulating beneficiation processes as a way forward in process mineralogy. The majority of process mineralogical studies can be described as optimization problems where conditions for the best economic and most resource efficient processing are sought. For solving such problems the raw process mineralogical data is not enough; optimal solution can be found only through modeling and simulation. Automated mineralogy provides accurate and detailed information on processed materials but this data is currently only partly utilized since the use of liberation information in modeling and simulation is very rare. He described the steps needed for establishing such simulations; from experimental work and liberation analyses via mass balancing, model development and model fitting to simulation of different scenarios.

This was followed by a number of papers describing the measurement and interpretation of mineral textures, a controlling factor in mineral liberation, and the use of such information in designing comminution and flotation circuits. Four of these were given by the 5-woman strong team from Australia's JKMRC, led by senior scientist Elaine Wightman, and including three graduates from the Philippines.

With the JKMRC team, Elanie Wightman, Tamsyn Parker, Kate Tungpalan,
Riza Mariano and Vannie Resabal

Will Goodall and Al Cropp

The coffee break gave us the first chance to take a look at the small exhibition and to talk to Al Cropp and Will Goodall who are co-chairing the morning session. They were both involved with Intellection in the early days of QEMSCAN technology. When Intellection went into receivership in 2008 its QEMSCAN assets were taken over by conference sponsor FEI.

On leaving Intellection Will formed his own company, Melbourne based MinAssist, a progressive young company offering services to the minerals industry in the interpretation of mineralogical information, and also a sponsor of this conference. Will is leader of the AMIRA P843 Geometallurgical Mine Mapping and Modelling Program (MEI Online).

Although based in Cornwall, UK, Al worked closely with Will as a consultant, but recently moved to Cambridgeshire to take up a position with Carl Zeiss, also a sponsor of the conference. Three months ago Zeiss launched a new automated mineralogy system for the mining industry, which is on display in the exhibition, and is described in MEI Online (see also posting of 18th November).

Igor Tonzetic of Zeiss demonstrates the Mineralogic Mining Automated Solution

During the afternoon session, a joint paper by Carl Zeiss, Australia's JKMRC and Anglo Technical Solutions Research, South Africa, described the use of X-ray computed tomography for 3-dimensional mineral characterisation, and a paper from the University of Utah also discussed 3D analysis of packed particle beds. This was presented by Chen-Liu Lin, and co-authored by Jan Miller, who presented an excellent keynote lecture on this subject at Process Mineralogy '12. As most people know, Prof. Miller was seriously injured last year in a car crash in Tibet, and was unable to travel to Cape Town this year, but it is good to hear that he is on the way to a full recovery.

Chen-Liu Lin (3rd left) with Ben Thompson, Teresa McGrath and Nicole Meyer

After a very long first day, which included 16 papers, it was good to relax in the Vineyard Gardens during the 'Happy Hour' (see also posting of 17th November).

Chen-Liu Lin with Roberto Galery, Otavio Gomes, Paulo Brandao and Marek Dosbaba

Tuesday November 18th
Acid mine drainage, a growing problem in South Africa (MEI Online), was the theme of the three papers this morning prior to the coffee break. In his keynote lecture, Bern Lottermoser, of Camborne School of Mines (CSM) UK showed that total worldwide liability associated with the current and future remediation of ARD is approximately US$ 100 billion. Consequences of failing to predict ARD for individual operations and for the mining industry include unplanned spending on remedial measures and reputational damage. Despite these severe risks, predicting the properties of mine wastes is typically not an attribute which is strongly embedded into the development of mineral resources, and examples of failures to predict waste properties accurately are plentiful. He stressed that the time has come to drastically improve our scientific efforts to forecast the likelihood of ARD accurately. Improvements in our predictive capabilities will come from new field and laboratory tests and the application of state-of-the-art characterisation tools and methodologies at individual exploration and mine sites. Such data are needed to establish the operational challenges, impacts and closure liabilities of ores and wastes.

There is reason for optimism that the required progress is possible. Such optimism is based on the phenomenal advances which have been achieved since the 1960s in our ability to predict ore and waste properties, processes and impacts. While the research community needs to establish practical state-of-the-art characterisation tools, industry has to accept and use such tools if we are to achieve more cost-effective mine closure and reduce environmental liabilities in the long term. Such progress also requires the application of predictive tools at the beginning of the life-of-mine cycle. At the earliest stage of mineral resource development, tests and methodologies should rely on integrated field and laboratory measurements using mineralogical, geochemical and geometallurgical tools. A more predictive approach to early ore and waste characterisation supports more effective management and valuation during operation and ultimately less costly mine closure outcomes.

Anita Parbhakar-Fox and Bern Lottermoser

Prof. Lottermoser divides his time between CSM and the University of Tasmania (UT) and he co-authored the 2nd paper of the morning, given by his post-doctoral student Anita Parbhakar-Fox of UT on the prediction of acid rock drainage from automated mineralogy. She showed how an accurate understanding of acid forming potential can be gained based on a detailed knowledge of mineralogy. A case study on the application of mineralogy in the interpretation of laboratory scale acid rock drainage prediction tests was then given by Megan Becker of the University of Cape Town (UCT). The study used a mineral mass balancing approach on a gold ore to understand the contribution of various minerals to laboratory scale acid base accounting and net acid generation tests, as well as on the novel UCT biokinetic test.

During the coffee break, I looked in on the Oxford Instruments booth. The company is showcasing a new version of their Automated Mineralogy Suite, INCAMineral, which now offers fully automated minerals identification and greatly enhanced reporting options on mineral liberation, deportment, particle and common perimeter associations, phase abundances, size distributions and element and phase grade recovery. Analysis options are available for modal or particulated samples and rare particle search functionality is integrated to increase throughput. They were also demonstrating new applications enabled by combining crystallographic (EBSD) and compositional (EDS) analysis.

At the Oxford Instruments booth

Following coffee were 5 papers on the use of XRD techniques, given by authors from Germany, Sweden, The Netherlands, Brazil and UK.  Oxford Instrument's INCAMineral and EDS were utilised in work described by James Strongman of Petrolab, UK, on the increasingly strategic REE bearing ores and the need to characterise and process these often complex ores. The case study looked at a flotation concentrate from a deposit in the Fen carbonatites, Norway. The fine grain-sizes and wide range of REE bearing minerals create a number of issues for classification by traditional automated mineralogy systems. INCAMineral combined with the next generation of large area silicon drift detectors allowed the particle detection and EDS acquisition to be optimised to produce an accurate, clean and robust data set, capable of detailed mineral speciation.  He concluded that INCAMineral is an extremely quick, powerful and cost effective tool for assessing complex metallurgical test products and correlating them directly with chemical assay.

Quantitative phase analysis using X-ray diffraction (XRD) has became a standard tool for process optimization and quality control in industrial environments such as mining or metals production. A presentation by Uwe König of PANalytical B.V. showed how Partial Least-Squares Regression can be used in industrial environments to easily and precisely predict properties such as crystallinity and process parameters more directly from hidden information in XRD data.

Uwe König (3rd left) at the PANalytical booth

Conference sponsor Bruker has two technical presentations this week and the second of them this morning showed how energy dispersive linear detectors and dynamic beam optimisation, recently available with Bruker XRD instruments, have opened the door to routine quality control XRD on poorly crystalline material such as clay.

At the Bruker booth

The afternoon session was an eclectic mix of papers from UK, Finland, South Africa and Australia, dealing with the process mineralogy of vermiculites, spodumene, sphalerite, porphyry copper, gold and PGMs.

Then after coffee, we were off to nearby Kirstenbosch Botanical Gardens for the conference dinner. Great setting, great company, great food and great entertainment (see also postings of 18th November and 19th November).

Wednesday November 19th

The last technical session of the conference contained a number of case studies involving mineralogical characterisation to optimise processing of various ores, including complex sulphides, nickel laterites, kaolin, heavy mineral sands, iron and gold.

Tantalum and niobium minerals (coltan) belong to so called "conflict resources" which are mined in civil war regions such as the Democratic Republic of Congo (DRC) and others in order to finance the fighting of rebellion groups.  A law was introduced in 2010 to prevent U.S. based companies trading the coltan minerals from DRC. Implementation of this law brought a need to characterize publicly traded coltan concentrates. Marek Dosbaba of Czech company Tescan Orsay Holding presented an interesting paper this morning on the TESCAN Integrated Minerals Analyser (TIMA), a system which could significantly help the characterization. The device provides information on proportions of minerals, which are precisely chemically characterized using a wavelength dispersive spectrometer without the need to move the sample out and without a new search for specific particles, allowing the very precise fingerprinting of ores from a particular locality.  Earlier in the month Australia's CSIRO installed a TIMA, the first in the country, a second due to be installed at Curtin University in January (MEI Online).

At the Tescan booth

The future of process mineralogy was the theme of the afternoon panel discussion, chaired by Megan Becker of the University of Cape Town, South Africa, with panellists Bern Lottermoser of University of Exeter, UK, Will Goodall of MinAssist, Australia and Elaine Wightman of JKMRC, Australia. The proceedings of the discussion will be reported in a later posting.

Amanda closed the conference, and invited delegates to attend Process Mineralogy '17 at the Vineyard in March 2017. A farewell wine function in the hotel gardens closed what has been a productive and enjoyable week in Cape Town, with interesting and friendly people. The atmosphere of the week is captured in this 7-minute video.

UCT's Prof J-P Franzidis was a welcome visitor to the farewell wine function

The Proceedings are available from MEI, and a special issue of Minerals Engineering, containing selected papers, will be published next year.