The Metplant ’13 conference started on July 14, with the GD Delprat Distinguished Lecture on Flotation given by Prof Graeme Jameson, Laureate Professor at the University of Newcastle, Australia, and one of the nominees to the International Mining Technology Hall of Fame. His lecture ‘Size matters- coarse and quick flotation can reduce costs’  discussed the everpresent need to reduce the costs of mining and milling operations. The greatest cost in ore concentration is the energy consumed in size reduction, particularly in grinding.

Some progress has been made in reducing energy consumption in grinding, through better use of existing technologies, and the introduction of grinding methods such as HPGR. However, most attention is usually given to the grinding operation itself, with little reference to downstream separation processes beyond a target grind size. Since flotation is widely used to separate the values from the gangue, the particle size distribution of the particles leaving the grinding circuit is generally determined by the known capabilities of conventional flotation machines.

Existing flotation machines work very well for sizes typically in the range of 50 to 150 μm. If the upper size limit for flotation could be increased, by innovations in machine design, there would be dramatic reductions in grinding energy, which would lead to savings of great consequence for the running costs of the whole operation.

In his talk, the effect of the final grind size from the grinding circuit on the energy costs for a typical base metal concentrator were discussed, with reference to a simple grind/float/re-grind/float circuit. Potential savings will arise not only from the reduction in energy costs, but also in the media costs that are of the same order. The talk finished with considerations of the way in which the flotation process could be improved, to increase the recovery of coarse particles, using new and innovative technology, such as fluidised bed flotation.

An important observation made by Jameson in his conclusions was that flotation researchers and comminution specialists should talk more to each other. He gave the example of MEI’s Flotation and Comminution conferences, Flotation’13 and Comminution ’14, which tend to have delegates dedicated only to each of these subjects.

Damian Connelly, Director/Principal Consulting Engineer, Mineral Engineering Technical Services (METS), presented a paper (Trends with selection and sizing large flotation circuits – what’s available in the market place) last month at COM 2013, the Conference of Metallurgists in Montreal. He reviewed the current state of the technology and flotation equipment including equipment suppliers and cited significant technical changes. “Flotation cells have increased in size over the last thirty years and there have been significant changes in cell layout, technology and design. The key factors for sizing and equipment selection are the ore character based on flotation test work and plant experience.” The retention time required at each stage is also a key factor he discussed. “The change from square Hough trough cells to cylindrical tank cells up to 600 m3 in size has been game changing. The recovery of a flotation cell relates directly to the amount of air added to the cell. Optimising reagent chemical use has provided great insights into flotation performance.

“Flash flotation cells, column cells, Jameson cells, unit cells, improved metallurgical performance, internal launders, etc. in stream analysis are all developments worthy of mention. Advances in modelling, simulation and control optimisation are bringing benefits to the industry.” He also outlined examples of large flotation cell installations and related issues.

He concluded that in telephone discussions with most of the suppliers, they agreed that circular tank type cells will be preferred compared to the square cell, “mainly due to the reduction of particle short circuiting in tank cells thus enhancing the possibility of recovery.

“The coarseness of the grind was a concern for some of the suppliers. To ensure particle suspension, solid concentrations of about 40% would be required. This will result in higher installed power per volume of cell which might limit the maximum cell volume.

“No definite answer to the preference of the two mechanisms, bottom or centre mounted can be found in the industry. Forced air bottom rotor and stator assemblies are currently more favoured in new installations and evidence was found where Wemco mechanisms were replaced by these mechanisms.”

FL Smidth has introduced a new mechanism into the market which sits in the centre of the cells but is also forced air.  “As an example when trying to float native copper, no evidence could be found that a particular flotation machine would be more suited to enhance native copper recovery in a flotation cell.  Several process or mineralogical aspects, e.g. particle size, liberation, floatability, reagent regime etc. will play a larger role in the recovery of the native copper minerals than the particular type of cell used. Some operations have installed competitor cells to run trials on their ore feed.

“Some cells are not suited for very coarse and high specific gravity feeds. Some vendors have lots of installations and maintenance history and personnel have their preferences particularly for new projects. People have their favourites and are reluctant to shift allegiances even for new projects.”

Flotation cell design

FLSmidth Minerals’ Asa Weber and Dariusz Lelinski explain that the company uses a continuous process improvement program to develop new flotation equipment and improve the performance of its existing flotation product line. “This process involves several disciplines; fundamentals in FLSmidth’s research department, senior design engineering in product development, collaborative work with academia, and guidance from the mineral industry.

“First principle models have shown that recovery of the mineral is a product of pulp and froth zone recoveries. By incorporating this principle, FLSmidth has been able to extend the boundaries of flotation equipment design to include machine characteristics that influence pulp recovery, froth recovery, and expert control of the flotation circuit.

“Historically, flotation cell design criteria and scale-up have emphasised pulp zone recovery.  Scale-up of Dorr-Oliver or Xcell forced air flotation machines and Wemco self-aspirated flotation equipment cells was accomplished by maintaining geometric and hydrodynamic similarity within the product line. The hydrodynamic functions are a set of dimensionless numbers which include Reynolds number, Power number, Froude number, superficial gas velocity, bubble air surface flux and air flow number.”

Computational fluid dynamics (CFD) models are now essential to the process improvement program. Stress and vibration analysis uses output from CFD models to provide the forces on impellers and stators. More importantly, CFD models have enabled the designer to understand a machine’s hydrodynamic characteristics and impact on bubble particle collection (pulp zone recovery). “Initially, the CFD model determines the spatial distributions of air volume fraction (also called void fraction), dissipation rate and flow fields within the vessel. Then in a post-processing simulation, the cell’s hydrodynamic characteristics are combined with first principle flotation models. The post processor simulation provides insight into the interaction between the cell’s hydrodynamic characteristics and bubble particle collection (pulp zone recovery).

“Properties that can be evaluated with the simulator include bubble size, particle diameter, specific gravity, hydrophobicity (contact angle) and surface tension. The actual pulp zone recovery rate will depend on the balance between the favourable effects of dissipation rate in increasing the collision frequency against its adverse effects on increasing the detachment rate. Therefore, knowing the spatial distributions of both, throughout the machine, is essential in understanding the effectiveness that different components (rotor, stator or disperser) have on flotation efficiency.”

The benefit to the designer is illustrated in the figure. Here the designer is investigating a Dorr-Oliver standard rotor against a new design. The machine design factors are the aforementioned cell’s hydrodynamic characteristics, and the controlled process properties are bubble size, specific gravity, contact angle and surface tension. The process variable is particle size. As shown, the location of fine and coarse particle collection not only varies for a given cell geometry, the new Dorr-Oliver stator design also expands the cell’s collection area.

Weber and Lelinski note that as the bubbleparticle matrix migrates upward in a froth phase, the bubbles interact with each other and become larger. As bubbles become larger, their surface area becomes smaller, restricting the number of hydrophobic particles that can be carried upward and flow into a launder. “Thus, the throughput of a flotation cell depends on bubble coarsening and residence time within the froth phase. Therefore, it is important to understand the basic mechanisms of bubble coarsening. At present, collaborative research between VA Tech and FLSmidth is ongoing to develop these relationships.

“For now froth phase recovery is recognised as a contributing, one could argue controlling, factor in a flotation cell’s metallurgical performance. If pulp recovery is such that the amount of material floating to the pulp froth interface is greater than can be removed at the surface, recovery is limited by the cell’s froth carrying capacity. When this occurs, there is insufficient bubble surface area to carry all of the floatable particles through the froth. While the carrying capacity restriction is often insignificant for smaller cells, it is of great importance in the design of larger flotation cells.  This is due to the fact that the specific surface area of the cell (ratio of the cross-sectional area to the volume) is much higher for smaller flotation cells. To accommodate for limited specific surface area, especially in high mass pull; coal, iron ore, cleaner applications, the design engineer uses external launders.

“Other design challenges imposed by large flotation cells are froth travel distance and froth residence time. For FLSmidth’s Wemco Smartcell and Xcell Flotation cells, this is accomplished with the installation of vertical baffles, radial launders and a froth crowder. Vertical baffles decrease the froth’s residence time and radial launders decrease froth transit time. Froth crowders for both the Wemco and Xcell flotation cells direct the slurry flow from the centre of the flotation cell to the periphery launder, which decreases froth residence time. All three design modifications increase the probability of bubbleparticle aggregates surviving the froth phase.

“Due to the fundamental difference in the hydrodynamic characteristics of the machines, FLSmidth’s forced air Dorr-Oliver flotation cells seldom use a froth crowder. Instead high efficiency radial launders have been designed. The entrance to the radial launder consists of a triangular inlet which maximises lip length near the centre of the vessel. The design also includes a steep angle inlet which maximises the momentum transfer of the froth into the radial launder. In scavenger application where barren froth exits, froth crowders are used in Dorr-Oliver cells to reduce the cell’s cross sectional area, which reduces froth residence time, stabilises the froth and promotes froth drainage thereby increasing both recovery & grade in this application.

“The development of very large flotation cells has also led to an understanding of particle dropback from the froth phase. Most bubble-particle aggregates have sufficient buoyancy to rise in the low-air fraction pulp. However, if particle size/density is high, the aggregate may not be able to rise through the high-air fraction froth. This phenomenon can have a dramatic impact in coarse particle flotation systems. To overcome this reduction in Pf, plant operators will operate their flotation circuit with very shallow froth depths. Of course, this can result in undesirable entrainment of fine gangue material in the concentrate.”

Responding to this, FLSmidth has developed the Froth Miner; an extraction system that vacuums material from select locations within the froth and separates the recovered solids into coarse and fine fractions with a hydrocyclone.  The Froth Miner is supported and located by an actuator connected to the cell’s level control so that the suction heads are maintained at an offset distance from the slurry level. This can increase the recovery of coarse material by selectively recovering the material close to the bottom of the froth where drop-back of coarse particles typically takes place.

The Froth Miner consists of four conical skimmer heads that are plumbed together and share a common actuator.  Froth Miner plant trials “have demonstrated an increase in overall recovery as well as the coarse particle recovery in scavenger flotation cells. With the automated level control, it is possible for recovery of specific froth depths to be maintained and controlled. Single cell recoveries of 15-36% copper have been demonstrated.

“FLSmidth’s cells incorporate design features to maximise both pulp and froth recovery.  However, large variations are often observed in the flotation process due to the inherent variability in feed characteristics. Thus, manual control by operators looking at the cell surface periodically and taking actions does not maintain stable operating conditions. FLSmidth’s Automation group has developed an Expert Control (RCS) package which incorporates the FrothVision system.”

The ECS/FrothVision system is an advanced image analysis system designed specifically for analysis of froth characteristics in flotation. It comprises all the necessary hardware and software to conduct froth image analysis and reports information relating to bubble size, bubble count, froth colour analysis, froth stability, froth texture and froth velocity which is used to assist the control of the process. The objective of the ECS/FrothVision system is to improve the operation and control of cells by taking advantage of image processing techniques.

The froth vision application records a sequence of images from each camera and calculates relevant features, such as colour, size distribution and mobility. The calculated images features will be stored in the ECS database, which can then be used for trending, alarming, optimising flotation reagent additions, and air control.

FLSmidth recognises that a flotation circuit’s performance is affected by both pulp and froth phase recovery. Weber and Lelinski conclude that use of FLSmidth’s “ECS/FrothVision system allows the entire flotation circuit to be optimised.”

Is bigger better?

Eric Bain Wasmund, Global Managing Director, Eriez Flotation Division (EFD) notes that “a major trend in flotation unit operations is to build larger flotation machines. Current practice dictates that as the head grade of ore diminishes, rougher flotation requires more capacity, which is economically addressed by using very large flotation machines. There are certainly many advantages to using fewer large machines including asset maintenance, lower specific energy, and a manageable plant footprint. In fact, the major equipment producers have worked hard to make their flotation technology platforms upwardly scaleable. However, there is a trade-off in which an increase in unit size is balanced against metallurgical performance and mechanical stability.”

He goes on to point out that what is often a secondary consideration “is whether the design is optimal from the point of view of recovering all of the different types of ore particles that are present. Consideration must be given to both the particle size and liberation distributions of the ore. Decades of research have shown that the basic mechanisms of particle collection in the pulp and froth zones of a flotation machine are different based on size and liberation of the ore particles. As a result, using large vessels as roughers is a compromise since some ore particle types are recovered more efficiently at the expense of other types.
This is shown in the graph above, which shows flotation recovery by size for a wide variety of mineral systems. This type of pass-band characteristic means that mineral recovery in the fine and coarse size fractions is often sacrificed to get good recoveries in the middle of the size range. Hopefully, that range coincides with the majority of the valuable mineral deportment.
Additional grinding cannot fully address this problem; while it reduces the fraction of unrecovered coarse particles, it simultaneously increases the amount of unrecovered fines. The net effect is simply pushing the size distribution from one region with low recovery to another.

As a flotation technology and equipment supplier, the EFD proposes another solution to this problem. In the case of fine and coarse fractions, there are different flotation technologies that are designed to achieve optimum recovery and grade for each size class – one type of flotation machine does not fit all applications, says Wasmund.

“For example, if the particle size is fine and liberation is good, a column cell using EFD’s proprietary spargers for producing a high air flux of ultrafine bubbles is considerably more effective. This high-flux solution can be accomplished using either the EFD low-energy SlamJet™ insertion sparger or the CavTube™ external, dynamic sparging system. The latter can also be used in a feed pre-aeration configuration to boost flotation kinetics. The CavTube uses the phenomenon of cavitation to selectively nucleate micron-sized bubbles directly onto hydrophobic surfaces. Both of these systems are effective for floating fine, well liberated particles. EFD has designed and built more than 700 flotation columns based on this family of technologies.

“If the size is coarse and the liberation poor, then EFD’s HydroFloat™ offers an effective solution. The HydroFloat technology uses an aerated fluidised bed that is effective for floating semi-liberated, coarse particles that are often missed in conventional flotation machines. The use of a dense phase, fluidised bed eliminates axial mixing, improves coarse particle residence time and increases flotation rate through improved bubble-particle interactions.”

“An effective way to use the most suitable technology for each size fraction is to split the feed using size classification such as cyclones or a teeter bed separator such as EFD’s CrossFlow™. This allows each size fraction to be recovered by best-in-class technology; thus reducing the amount of recoverable ore that is misplaced by conventional flotation cells that process feed with a broad size distribution. Split feed arrangements are being used commercially for phosphates, potash and coal and are being developed for base metal sulphides.”

He concludes that “by carefully considering the ore characteristics, innovative flotation circuits have been designed and operated that take advantage of the most suitable flotation technology for each major size class. This is a process design philosophy that rubs against the idea that bigger is better, especially if it enables better metallurgical performance.”

Flotation testing

Since the publication at the SME1 describing XPS Consulting’s practice of High Confidence Flotation Testing (HCFT) and how this reduces the project scale-up risk, which was reviewed by IM (July 2009, p21), XPS has further advanced its flotation testing practice by integrating the minimum sample mass and safety line models of Gy2 into the HCFT practice. Other new sampling models for drill core developed by Oliveira3 have successfully addressed the typical problem of limited availability of sample material. These ore samples are studied using a microprobe and the modern FEG QEMSCAN to produce highly detailed mineralogical information from which powerful metallurgical processing implications are formulated. Because the sample material is truly representative, the mineralogical data and processing implications are sound.

On the flotation testing platform, the initial grinding strategy uses the mineralogical data and saves both time and precious sample material by engaging appropriate grinds in the first tests. This avoids the older empirical practice of hunting test-by-test for the ideal grind. The XPS Mixed Collector system then selects candidate collectors from the mineralogical data, again saving time and sample material, resulting in accurate, more rapid flowsheet development delivering better grades and recoveries. The continued use of HCFT produces tighter metal balances now as a result of the improved sampling and subsampling practice.

This new approach was used on the development of the flowsheet for Ivanplats’ Kamoa copper project, Dr Norman Lotter, Consulting Metallurgist explains, “and resulted in a Milestone Flowsheet producing viable concentrate grades and recoveries that warranted the launch of the project prefeasibility study after only ten months of laboratory testwork4. The total error in the copper metal balances averaged -0.3%. In this case, the hypogene ore carried a range of copper sulphides including chalcopyrite (26.3%), bornite (50.9%), and chalcocite (17.1%), with minor covellite, occurring at a very small mean grain size of 7-27 μm. It would be uneconomical to grind the ore down to these sizes to deliver good liberation before flotation. Rather, a compromise of under grinding at this stage together with a strategy to float both middling and liberated copper sulphides was formulated.

“Mixed collector suites, when optimised, are better at the flotation of middling particles than are single-collector suites. Accordingly a mixed collector suite was essential for the successful flotation of this range of sulphides, because these sulphides also divide electrochemically in semiconductor characteristics that require different potentials to float. The mixed collector system offers a mixed potential platform, thus the first optimised collector suite used was a diisobutyl dithiophosphate and an isobutyl xanthate, in the mass proportion 36:64%. This proved to be successful at floating liberated and middling particles to rougher and scavenger concentrates from the primary circuit.”

XPS has also been developing its capabilities in the flotation of non-sulphides, including rare earths. Non-sulphide flotation tends to rely upon physical absorption resulting from electrostatic potential or hydrogen bonding rather than chemisorption. The physical conditions that need to be controlled can include pH, temperature and water chemistry (especially the presence of mobile cations). Sometimes organic chelants are used as pre-conditioners to sequester ions such as calcium or iron from mineral surfaces and from solution. Particle-particle interactions can also be significant, especially those involving sliming of small charged particles and displacement of collector on pay minerals. Thus, non-sulphide flotation circuits commonly employ desliming prior to flotation, or pre-conditioning with dispersants. The collectors for non-sulphide flotation are selected according to ionic charge in solution, functional groups, hydrophobichydrophilic balance, stereology, and sometimes chelating properties. The complexity of the number of variables in non-sulphide flotation results in expertise being partly theoretical and partly empirical, whereas the circuits themselves commonly employ a sequence of conditioning stages prior to flotation to achieve a commercially successful separation.

Managing the process

Outotec has a major installed base at FQM Kevitsa – a nickel-copper mine in Finnish Lapland which will soon be featured as an IM Operations Focus. The Outotec delivery to Kevitsa included three grinding mills, flotation cells, thickeners, filters, samplers, analysers and automation systems. It delivered two AG mills, one pebble mill and 73 Outotec TankCells ranging from 20 to 300 m3 cells and a Courier 6i elemental analyser for the flotation circuit for froth control along with 36 FrothSense imagers and the Outotec ACT (Advanced Control Tool) for process optimisation in flotation.

The ACT system with 36 FrothSense imagers – one imager for each TankCell – controls and optimises the process. Courier and FrothSense enable constant froth analysis based on imagers and froth speed. The ACT system can stabilise the speed and regulate the froth level and speed in different cells, as well as run the process close to the low grade – when the recovery improves.

“Courier and FrothSense are great tools for analysis and control, but when combined with the ACT they enhance the concentrate and enable the operators to better control and supervise the overall process while maintaining the recovery at the optimal level,” explains Niko Koski, Project Manager for Outotec.

“FrothSense has worked well,” confirms Anthony Mukutuma, Plant Manager at FQM Kevitsa Mining. “We have here 73 cells, and we need to make sure that all the cells and rates are working well. With FrothSense imagers we get a full and continuous view. It is difficult to predict the process, and operators can lose the edge and motivation to constantly monitor and adjust. The
ACT keeps us alert all the time and on top of what is happening in the floats. FrothSense and the ACT system is like having an operator alert and at full attention every second.”

Along with the process equipment delivery, commissioning, installation and ramp-up services, Outotec has also provided Kevitsa operators with equipment training.  “Outotec’s Virtual Experience Training provides customers with a fast and effective way to train their personnel already before the production start-up phase and later according to individual customer needs,” says Kai Rönnberg, Team Leader in Product Management at Outotec.  “Virtual Experience Training provides hands – on experience with operations by utilising an advanced training simulator developed by Outotec. In the training we teach the theory of minerals processing but also provide simulator exercises focusing on the basic operations of the equipment and process circuit and troubleshooting scenarios. The target of the training is to show operators how to optimise the process circuit and give advice how to maximize the recovery and profit in changing conditions,” he continues.

“VEX training is a very good tool for providing operators with an intensive and fast way of simulating the real environment and how the actions taken running the process impact on performance (grade and recovery). VEX is good package and we got the result we wanted,” Mukutuma adds.  “I participated in the training course, and it is a good simulation even for experienced metallurgists to get to know the automation system before the start-up.”

Outotec has signed service agreements with FQM Kevitsa for the Outotec Analysers, PSI 500 and Courier 6i with the Advanced Control Tools (ACT) and Outotec Integrity. The aim of these agreements is to keep the specialised equipment available and in good condition.

Proper flotation cell level control is one of the most important yet overlooked parameters in a mineral processing plant, especially in new plants, says XPS Process Control. Upstream flow disturbances, improperly sized actuators, poor controller tuning, including dead time between control actions and responses, and improperly selected/commissioned instruments all contribute to the problem in maintaining cell level – a key control of one of the main concentrator KPIs (key performance indicators) …concentrate grade and metal recovery.

The XPS Process Control group has many years of experience in all aspects of flotation cell level control optimisation and recently, in collaboration with global instrument supplier E+H (Endress and Hauser), is marketing what it describes as “a best practices solution” – a unique and simple sensor device to measure flotation cell level accurately and consistently.”

XPSFloat™ consists of a conical float device made of a robust, self cleaning material combined with a float target and, typically, an E+H (non contact) ultrasonic measurement sensor. The device floats on the pulp/froth interface while the level sensor and target work in tandem to measure the cell level, accurately all the time. The level signal is delivered in real time to either the plant DCS or PLC (control) system which is often paired with a control loop to change actuator position and maintain a consistent pulp level. XPS Process Control says “the practice is not new, but XPSFloat is unique in its ability to maintain reliable, trouble-free measurements in this normally rugged environment. Better measurements lead to better control and are a requirement for optimal metallurgy.

In the Cerro Bayo district of Southern Chile, Mandalay Resources operates gold-silver mining operations and recommenced mining and processing and subsequently shipped its first concentrate in February, 2011. Ramp-up to the currently planned 1,200 t/d of ore was completed in the fourth quarter of 2012 with planned expansion to 1,400 t/d by the first quarter of 2014.

Mandalay Resources was looking for instrumentation upgrading assistance to improve its flotation cell level control and selected Sedgman, Chile (an EPC/EPCM company), to help together with Endress+Hauser, Chile Ltda.  Following stand fabrication, with a local fabricator, and with the support of Sedgman the units were installed and commissioned on the plant flotation cell.

Additional XPSFloat components have also gone to Europe and to Australia through the E+H office and sales network.  Jack Evans of Hawk Measurement America and Robert Stirling, Hawk Measurement Systems in Australia explain that “optimisation of flotation cells is a constant need for profitability in the concentrator. Most, older cells use a displacement float below the froth layer to measure the pulp height.

“Pulp height is an extremely important process measurement, used to ensure that liquid pulp is not allowed to overflow to the launders. If pulp overflows, the flotation cell ceases to function effectively, or if the pulp is too low little material will flow over the launder, which are both very costly to the process. The traditional displacement float technique has proven to be limited in performance in a variety of ways: the float may at times stick, slurry builds up on the float mechanism changing the effective specific gravity tracked, and the floats can be affected by high agitation, etc.

“The development of a very low frequency Acoustic Wave Transmitter has changed the way cells are controlled. Three non contacting measurements can be made to control the pulp height, foam height and foam density, without contacting the process media.

Evans and Stirling say a displacement float (an intrusive device) floats on the liquid surface below the froth and the position of this device is very dependent upon the liquid media density.  “This system becomes inaccurate when air velocity increases (aeration) causing a change in the discernible froth and slurry interface.  Subject to scaling and jamming. Requires periodic cleaning and as mentioned subject to the certain and continuous changes in pulp density.”

Hydrostatic Pressure Transmitters (also intrusive devices) measure the hydrostatic pressure change as the slurry level changes.  “Accuracy is affected by density change in the slurry. Also affected by scum build-up and requires periodic cleaning and calibration.” Multi probe conductivity (intrusive) measures the difference in the conductivity, which is dielectric dependant, and the level between froth and slurry. “Intrusive probes suffer with scale build-up and changing dielectric levels between the froth and slurry. Requires constant calibration.”

However, they say that a non-intrusive low frequency acoustic device measures the impedance change between the froth and slurry.  [It] is “not affected by density dielectric or scaling.”

Hawk’s Acoustic Wave Transmitter “will penetrate through the froth to measure the pulp height. The sensor is mounted above the froth and pulp height, so it has no maintenance or mechanical problems. Typically the transmitter can be mounted at walkway height for easy serviceability. The low frequency level transmitter can be supplied ready for connection to the typical two-wire loop power supply used for the displacement float transmitter which it is replacing. Remote mounted transmitters are also an option.

“Hawk also provides as an option, a nonintrusive transmitter to measure the froth height.  Continuous measurement of the froth height, provided as feedback to the control loop for the inlet ‘Dart Valve’, allows a cell to maintain constant overflow of froth to the launder, even when the orebody type may produce variations to frothing consistency. Small changes in the pulp height to keep the froth overflowing at all times will increase the efficiency of the cell. Hawk transmitters will reliably measure froth height, even when froth density changes.

“Hawk also provides a third type of transmitter to give an indication of relative froth density.  Higher density froth will have greater entrainment of mineral going over the launder. Currently, density measurement is not widely used due to the degree of difficulty in making an effective on-line density measurement in each flotation cell. Bubbler type pressure transmitters have been commonly used, though they have high maintenance costs due to their intrusive installation. A non-intrusive transmitter that penetrates partly through the froth gives an output related to density. Data from the froth height transmitter is used with the froth penetration (quasi density) information.  Monitoring of the deviation between froth height and froth penetration allows the control system to track relative froth density – all non-intrusively. Relative density data can be used to actively control density through a feedback loop, regulating forced air flow into the flotation cell.  Air input is currently largely controlled manually by on site operators.

“To effectively measure each layer in a flotation cell the correct non-contact transducer must be used. Each has a different purpose but the technology is the same. Pulp level measurement requires a low frequency sound wave (5 kHz) to penetrate the froth layer without the sound wave being attenuated. Froth level measurement uses a higher frequency (20 kHz) to reflect the sound wave off the top of the bubbles. Density measurement uses a midrange frequency (15 kHz). Transmission of high powered acoustic waves ensures minimal losses through the environment where the sensor is located. Due to the high powered emitted pulse, any losses have far less effect than would be experienced by traditional ultrasonic devices.  More energy is transmitted hence more energy is returned. Advanced receiver circuitry is designed to identify and monitor low level return signals even when noise levels are high.  The measured signal is temperature compensated to provide maximum accuracy to the outputs and display.

“Testing for the Acoustic Wave solution is on-going with favourable results. Flotation cells differ in size, flow, orebodies, reagent addition, air flow, froth density, bubble size, etc. Testing has been completed on several different manufacturers’ flotation cells and varying orebodies and metal concentrations.  Current results are good and the testing will continue. Above [are] results from a nickel concentrator.

“Hawk’s low frequency transmitters require no maintenance due to their self-cleaning nature.  The high powered acoustic wave being transmitted will automatically clean the sensor face with every measurement pulse. Self-cleaning minimises build-up on the sensor facing which would otherwise prevent the sensor from measuring accurately.”

Reagent advances

Frank Cappuccitti, President, notes that over the last decade, “Flottec has conducted ongoing research with distinguished partners that has lead to some advancement in flotation technology. Initially, working in conjunction with Professor Jim Finch as a participant in the McGill Flotation Technology Chair, our work focused on understanding the fundamentals of flotation cell hydrodynamics and how flotation reagents affect the main hydrodynamic parameters such as bubble size, water recovery and gas hold up at various air rates.

“We learned that all frothers reduce bubble size and create froths as concentrations increased. But we also learned that the relationship between bubble reduction and froth creation was different for all frother chemistries.  All frothers tended to reduce the bubble to about the same size and that the concentration that this occurred is called the CCC or critical coalescence concentration. Therefore, a frother could be characterised by its hydrodynamic curve that defined its CCC (at a given air rate) and the amount of froth that it created at its CCC. Strong frothers created lots of froth while weak frothers created small amounts of froth at the minimum bubble size.”

“This lead to a new approach to frother optimisation in a plant, where a set of differing strength frothers are now added at their CCC for the cell conditions and the frother strength is changed until the froth conditions in the plant are optimal. This is done in a plant and again emphasises why frothers are very hard to optimise in a laboratory because it is very difficult to scale up the required froth characteristics needed in the plant from the lab cell. A froth that works in the lab may not work in the plant. This methodology has already been used in many plants resulting in better performance.

“In the last several years, as a result of the new understanding of flotation cell hydrodynamics and how they are affected by reagents, the research emphasis has now switched to the improvement in real time measurement of hydrodynamic parameters such as gas hold up to affect better control of the flotation circuit. Flottec is currently working with Cidra to test an online gas hold up measurement device to determine if gas hold up can be used as a control variable to optimise circuit performance. This research is ongoing and results to date have been very positive.”

Another Flottec initiative has been to provide much more training. As a result of the expansion of mining and lack of experienced metallurgists, Flottec found that in order to implement new ideas, it was very important for operators and metallurgists to better understand flotation basics. “Our customers supported the research efforts undertaken but in the short term, most likely benefited more from training programs,” Cappuccitti believes. “They realised that to undertake new approaches, first a basic understanding was required by operator, manager and metallurgist alike.

“With all the new capacity and new start ups in the last few years, it has also become apparent that our methodologies for designing initial mill reagent schemes are inadequate. Far too many plants are starting up with design reagents schemes that do not make sense. Once the plants start up, reagents often need to be changed completely. This is not the fault of the flotation design or engineering companies. It is partly because too little investment is made in flotation studies prior to start up. Also, most of the work is done in the feasibility stage where the ultimate objective is a study to prove financial viability and not to optimise performance. But this doesn’t mean we cannot improve on the current methodologies used to design a standard flowsheet for a new mill.

“Part of the problem with screening reagents is that there are too many collectors and collector blends to do proper screening. Also, even though frothers cannot be optimised in the lab, there are ways in which frothers can be used to allow better scale-up of the reagent scheme.  Flottec has developed a collector screening program for use in the initial phases of mill design. It uses pure collector chemistries as a basis for the initial screening. A candidate collector from each of the eight or nine families of sulphide collectors is tested in its pure form with no dilution with frothers or blending with other collectors. This will establish the activity of each collector family for each mineral in the ore. A frother is chosen again by adding at the CCC and getting the right strength to provide maximum kinetics. This ensures that it is the differences in the collector performance and not hydrodynamic factors that are being tested. Once the collectors are screened and the best chemistries identified, then optimisation work can be done in the next phase. This methodology can also be used in any plant that would like to re-evaluate its reagent scheme.”

Cappuccitti concludes that “based on new understanding of reagents and flotation cell hydrodynamics, better training and optimisation methodologies, we will continue to test new approaches in the plant that look at optimising operating strategies using air/mass and froth recovery profiling, and circuit control using gas hold up measurement. This will hopefully lead to the next level of improved metallurgical performance.”

Huntsman Performance Products has developed The POLYMAX®T10 and POLYMAX T12 low molecular weight liquid dispersants that have been shown to improve mineral recoveries and concentrate grades in the flotation of copper, copper gold, gold, carbonaceous gold, nickel, phosphate ores and coal.  The company says “the need to produce economically viable concentrates from more complex ores is driving the development of dispersants and depressants that can efficiently deal with gangue species such as clays, kaolin, magnesium oxides and silicates. These gangues reduce recoveries by inhibiting the interaction between collector and mineral, lower concentrate grades by reducing froth drainage and increasing gangue entrainment and lower process throughputs by forcing operations to lower

pulp densities to counter increased pulp viscosities.”

The most common gangue depressants are the high molecular weight polymers such as guar, carboxy methyl cellulose (cmc), dextrins, and their chemically modified versions. Polyacrylic acids, polyacrylates and alkyl sulphonates are also commonly used. These depressants require moderate to high dose rates (200 to 500 g/t) and are supplied as powders due to their high viscosity in dilute solution and degradation in aqueous solution.

POLYMAX T10 and T12 are non-ionic polymers of polyoxyethylene and polyoxypropylene. They are lower molecular weight liquids that disperse readily in water. They have both dispersive and froth modification properties that contribute to their effectiveness in improving mineral recoveries and gangue rejection. The effective dose range is typically 50 to 100 g/t.

These reagents are not regarded as a replacement for the more commonly used high molecular weight polymer depressants, but their dispersive efficiency and beneficial froth modifying properties makes them a useful component of an effective reagent scheme. They have been effective in improving mineral recovery and concentrate grade of ores containing fine and fibrous particles and clays and they can be used in highly saline water.

In the flotation of a fibrous nickel ore, for example, “flotation recovery of nickel was found to increase by 18% without loss of nickel grade with the addition of POLYMAX T10. Similar results were also obtained in the scavenger-rougher flotation circuit of a phosphate mine, showing a 17% increase in the scavenger-rougher P2O5 recovery and 11% MgO rejection improvement in the scavenger cleaner concentrate, compared to the baseline results with no dispersant added.”

In response to the industry’s need for safer, sustainable alternatives to NaSH, Na2S, and Nokes reagents, Cytec has developed AERO®7260 HFP reagent, which it says is “a highly efficient, selective, economically viable sulphide mineral depressant for copper suphides and pyrite with wide applicability.”

NaSH and similar reagents generate high concentrations of toxic, flammable, hazardous, and even lethal H2S gas which pose significant health and safety issues for plant operators and local communities. In addition, transportation of
20 to 40 t/d of 40% NaSH solutions presents shipping and handling hazards. NaSH metallurgical performance is also not robust as noted by large performance swings which accompany changes in ore mineralogy and poor pyrite depression even at very high dosages of NaSH.

Dr Mukund Vasudevan, lead R&D Manager of AERO 7260 HFP at Cytec’s Stamford Research Labs states that “customer plant trials confirm 7260 HFP as not only a safer alternative to NaSH but [it] also provides distinct performance advantages”. It is a polymeric depressant which, at just 0.25 to 1.0 kg/t, allows the replacement of 50% to 90% of NaSH. It is a stable and chemically inert reagent classified as non-hazardous to the environment, does not produce H2S gas, and can be stored and transported safely. AERO 7260 HFP provides operational and economic advantages while maintaining metallurgical performance under standard process conditions. Additionally, it eliminates pre-treatment of bulk Cu-Mo concentrate with steam, acid and CO2 conditioning and attrition conditioning.

Cytec lab studies and plant trials with AERO7260 HFP on a North American Cu-Mo concentrate demonstrated excellent Cu and Fe depression and Mo selectivity even after reducing NaSH consumption by 80%. The required AERO7260 HFP dosage was less than 0.5% of original NaSH dosage. Similar promising results have been obtained on several other Cu-Mo substrates as well. These results suggest that it is highly effective in the depression of both Cu and Fe and enables significant reductions in NaSH consumption. Clearly, the benefits of its use were confirmed by the improved metallurgical performance and substantially reduced dosage of NaSH.

Vasudevan also notes “AERO 7260 HFP can act in the rejection of gangue from sulphide concentrates and as a depressant of all
sulphide minerals while floating nonsulphide gangue,” for example in Ni-talc separation. “It is an innovative solution
for the flotation industry.”

On the move

Clariant, a world leader in specialty chemicals, has opened the new global headquarters for its Oil and Mining Services business unit in The Woodlands, Texas.  A centre of technology innovation just north of Houston, it expands on the company’s investment strategy in North America.  The campus includes a regional mining technology centre and a customer and employee training facility. The facility, which houses more than 100 offices, will serve all three parts of the OMS operations – Oil Services, Refinery Services and Mining Solutions.

“Clariant’s strategy is based on four pillars which are increased profitability, the fostering of innovation and R&D, intensified growth, and the re-positioning of the portfolio. The opening is a clear investment in the continued growth of our company,” said Hariolf Kottmann, CEO of Clariant.

The new Mining Technical Centre, equipped with a two-story flotation column, will focus on flotation chemicals, emulsifiers for explosives and fertiliser additives.  “The state-of-the-art laboratories will allow staff to cross-train, share strengths and fully engage in their roles giving us unique new assets and capabilities that set a new standard in the industry,” said John Dunne, Senior Vice President and General Manager of Clariant OMS.     IM