Search Documents

By J. Andrew Cornier, Mike Cooper

"Upon successful completion of the AG/SAG Mill project (Cooper et al., 2001) the Brunswick Mine concentrator was configured with single line processing through the AG/SAG mill, followed by two parallel ball mill and flotation circuits. The Stage Grinding project was undertaken to utilize a pre-flotation stage on the AG/SAG mill primary cyclone overflow and simplify the flowsheet by consolidating the two ball mill and flotation circuits, resulting in single line processing fi-om AG/SAG mill to final concentrate. Concurrently, the reclaim water supply capacity was increased to enable flotation cleaning at lower percent solids.Benefits from the Stage Grinding project include revenue enhancement from increased zinc grade and recovery combined with additional lead recovery. Further benefits included a reduction in the number of process pumps, reduction in loss associated with spillage to sumps and process simplification through efficient use of equipment layout.The Stage Grinding project was implemented in six phases over a period of seven months in 2000. Each phase was commissioned during scheduled concentrator shopdays with minimal additional downtime required. One year after project approval the Stage Grinding project was completed at a total cost of $1 1.8 million.INTRODUCTIONThe Noranda Inc. Brunswick Mine, located near Bathurst, New Brunswick, is one of the world’s largest leadzinc mines. Milling of ore began in 1964 from the No. 12 orebody, with an expansion in 1966 to accommodate ore fkom the No. 6 open pit located several kilometers away. In 1980 the pit was exhausted and the concentrator rationalized to accept a design feed rate of 10,500 tomes per day from the No. 12 mine. In excess of 100 million tomes have been milled to date, representing over three-quarters of the known reserve. The concentrator employs differential froth flotation to produce lead, zinc, copper and leauzinc bulk concentrates. The ore is massive sulphide, with approximately 20% valuable minerals, 60% pyrite and 20% silicates."

Jan 1, 2002

By James Scott

Outline Deposit Overview and Brief History ?Geologic Setting ?Local Geology ?Petrography ?Geochemistry ?Mineralization ?Geochronology ?Conclusions Deposit Overview ?Porphyry Cu-Mo-Au ?Discovered 1957 (BIK Syndicate) ?Various programs, most significant: ?Hecla(1970s) ?Teck(Late 1970s Early 1980s) ?Copper Fox Metals (2005 -Ongoing ?optioned from TeckComincoin 2002) ?Measured and Indicated Resources: ?847Mt grading: ?0.292% Cu ?0.019% Mo ?0.202g/t Au ?1.796g/t Ag ?Total ~ 0.485 % Cu Eq.

May 1, 2006

By N. M. Soonawala

Detailed regional geophysical and geochemical surveys, ground surveys and drilling were undertaken to follow-up the federal-provincial Uranium Reconnaissance Program (URP) over the 27,OOO-km2 Kasmere-Munroe area in northwestern Manitoba. An analysis of the URP data showed a fair correlation between the regional airborne radiometric results and lake-bottom geochemical results, both of which also correlated with the regional geology. More detailed surveys defined anomalies with greater precision within the broad zones outlined by the URP. The ground follow-up indicated that URP anomalies can be caused by varying degrees of uranium concentration in the rocks. Mineralization of interest was found in leucogranites at several locations within the survey area. Anomalous uranium concentrations were also found in coarse pegmatites and Hudsonian granitic intrusions. It can be concluded that large-scale regional surveys can be followed up in relatively short periods of time to yield meaningful results if a properly planned exploration sequence is employed, each stage of which successively narrows the area of interest.

Jan 1, 1979

By Seppo Haapamaki

"Since it started production in 1970, Anderson Lake Mine has experienced several radical changes in mining methods as a result of the physical characteristics of the ore zone along with the need for improved productivity. This paper examines the transition from longhole open stoping to mechanized cut-and fill then to vertical crater retreat at Anderson Lake Mine. The reasons for the need to change mining methods is analyzed taking into consideration the impact of such criteria as improved productivity and ground conditions and how these criteria affected the over-all economic viability of the mine. IntroductionAnderson Lake Mine is one of five producing copper mines in the Snow Lake area and is situated approximately 3 km southeast of the townsite of Snow Lake (Fig. I).The mine production rate at Anderson Lake is 700 tpd and all ore is trucked from the mine to the mill located approximately 2.5 km from the mine site . The mine is serviced by a three-compartment shaft sunk to a depth of 3300 ft with level spacings at 175 ft and 150 ft intervals from the 600 ft level to the 3000 ft level. .The shaft was collared in 1965 and, a year later, the hoist and compressors were installed. By 1969, a crusher system had been built and actual production began in 1970. During 1978, production was suspended for the year to allow for 430 ft of shaft deepening."

Jan 1, 1988

"Air Products approached LAC Minerals in 1987 to initiate trials with pure oxygen in the leach circuit at several LAC facilities in Canada. This paper gives an overview of oxygen use and covers specific operating data from the Doyon and Terrains Aurifères mills near Val D'Or in northwestern Quebec. LAC also uses oxygen at the Macassa mill in Kirkland Lake, Ontario.In addition to the LAC facilities, oxygen is currently in use for test purposes at the Holt McDermott mine in Kirkland Lake, Ontario and the Camflo mine near Val D'Or.There are two requirements for oxygen in the cyanide leach mill. Firstly, oxygen is required in the familiar gold dissolution reaction as follows:(1) 4 AU (s) + 8 NaCN (aq) + 02 (aq)+ 2H 20(aq) - 4NaAu(CN) 2(aq)+ 4Na OH(aq)Of the reagents on the left side of the equation, the availability of gold is fixed by the ore in question. Obviously, water is available in excess, 1eaving cyanide and oxygen as the limiting constituents in the process. Most mills operate with an excess of cyanide, which is fairly closely monitored. Dissolved oxygen levels, however, are not normally watched very closely and find their own level based on the temperature, the availability of compressed air and the ability of the ore to consume oxygen.Typical D.O. levels are in the 3-8 ppm range, however levels less than 1 ppm are not uncommon for refractory ores.With pure oxygen, dissolved oxygen levels as high as 30 ppm can be achieved. (Figure #1). This greatly accelerates the kinetics of the leach reaction and may also shift the reaction further towards completion thus increasing gold dissolution slightly."

Jan 1, 1990

"4.1. IntroductionThis chapter reviews the geochemistry of the platinum-group elements (POE) in sulphur-poor silicate rocks, and the limited data on POE in silicate, oxide, sulphide, arsenide and sulpharsenide minerals. The earliest geochemical studies of POE in silicate rocks, minerals and meteorites were carried out mainly by Goldschmidt, the Noddacks and their co-workers, and are reviewed in Crocket (1969). These researchers established the siderophile character of the POE, emphasized the common association of the group with sulphides in ultramafic-mafic rocks and suggested that very low concentrations of POE are typical of sulphur-poor silicate rocks and minerals. Much of these data were for Pt and Pd only and are more than 50 years old. Because of the poor sensitivity typical of the available analytical methods, few reliable data were obtained on silicate rocks in these early studies. Surveys of POE geochemistry demonstrate that the data base for sulphur-poor silicate rocks is very restricted, although considerable highquality data for Pd and Ir have been accumulated (Wright and Fleischer 1965, Yushko-Zakharova eta/., 1967, Mertie 1969, Crocket 1969, 1979, Keays 1981).Two recent publications, Vol. 71, No. 7 (1976), of Economic Geology (An Issue Devoted to Platinum-Group Elements) and Vol. 17, Part 2 (1979), of The Canadian Mineralogist (Nickel-Sulphide and Platinum-Group-Element Deposits), are of special interest. These volumes are clear testimony to the impressive progress made in the last decade toward describing the mineralogical and geochemical aspects of POE ore deposits, but at the same time demonstrate the constraints in modelling POE ore genesis due to our meager knowledge of the. POE content and distribution among minerals in the source rocks of ultramafic and mafic magmas. It is clear that the phases hosting POE in mantle source rocks are poorly understood. Nevertheless, Naldrett and Duke (1980) have shown that the absolute and relative concentrations of POE in magmatic sulphide ores are related to the degree of partial melting of mantle peridotites, and subsequent processes of batch equilibration and fractional segregation of sulphides."

Jan 1, 1979

By W. S. Dyer

The lignite deposit at Onakawana on the Abitibi river is of particular interest at the present time, since the Ontario Government has recently decided to proceed with the extension of the Temiskaming and Northern Ontario railway to James bay. Forty-five miles of steel will be laid this year to the crossing of the Moose river. This will cross the lignite field. It has been known for many years that lignite occurred in the Moose River basin, but it was not until drilling was commenced at Blacksmith rapids (or Onakawana, as the camp is now known) in June, 1929, that a seam of economic thickness and extent was found. Early in 1929 the Ontario Legislature voted $25,000 for the carrying out of a drilling programme in the Moose River basin. This came as a culmination of two summers' exploratory work by the Geological Branch of the Ontario Department of Mines, and as the result of a report which said; in effect, that further information concerning the mineral resources of the basin could only be obtained by drilling, and that such a drilling programme was well worth while in order to establish once for all whether lignite, fire-clay, or oil or gas did or did not exist in economic quantities in the region.

Jan 1, 1930

By Turgut Yalcin

"Recent years have seen a resurgence of interest in flotation cell design, and one of the areas where attention is being focused is flotation in centrifugal fields which aims at increasing the flotation rates by forced bubble-particle contact.A new cycloflotation device was developed at Laurentian University and tested on the nickel ore of Falconbridge Limited in Sudbury. In comparison with conventional flotation, cycloflotation yielded at least seven times faster flotation rates in terms of the mass of solids floated per unit time per unit flotation volume, while producing concentrate grades similar to those of conventional flotation. These results, which were obtained with a bench-scale ""cyclofloat cell,"" suggest that cycloflotation has the potential of setting a new direction in flotation technology.IntroductionFroth flotation is no doubt the dominant separation technique in the mineral processing industry. Since its inception early in this century, much research effort has been and is still being expended on this process. Until recent years, emphasis seems to have been placed on the chemical aspects of flotation, but there is now an increased awareness for the need to improve flotation machine performance and more and more attention is being directed towards this area. In a 1982 article on ""Flotation Machines"" (Young 1982) it is stated that "" ... by now the design of the machines used would have reached a high level of perfection, ... In fact, this is not the case ... "".The process of flotation in a flotation device may be considered to take place in several distinct stages as follows:"

Jan 1, 1994

By C. H. Lee, K. Adham, D. Lemieux, J. P. Mai, A. M. S. Hussein, S. Francey

Copper concentrates contain an increasing amount of arsenic, as the older sources of cleaner ores are depleted. At the same time, the market-determined price penalties on arsenic rich concentrates are rising due to limited arsenic treatment capacity at smelters and increasingly stringent environmental emissions limits, necessitating arsenic removal from the copper concentrates. Heating the concentrate to about 700°C can remove significant amounts of arsenic (as elemental and/or sulfide); however, it is important to achieve that removal economically and with due consideration to post-roasting arsenic (toxic waste) treatment. A comparison is made among different roasting options (kilns, multi-hearths and fluid beds) to qualify their potential application to arsenic roasting of copper concentrates. Both direct and indirect methods of heating as well as inert, air and oxygen atmosphere options are considered. Technology comparisons are made with respect to the roaster’s operation, its off-gas treatment and arsenic stabilization. INTRODUCTION In the last twenty years, the arsenic content in globally produced copper concentrates have risen steadily due to the depletion of cleaner ores and the increase in complex concentrate production. Smelter contracts typically limit arsenic to less than 0.5wt%, with a penalty paid per unit of additional arsenic and other contaminants such as antimony. While penalties can vary, they can be on the order of $70/dry metric tonne of concentrate per 1% of arsenic. Strategies used by miners to avoid the price penalties have been to blend clean and complex concentrates, and to remove arsenic from complex concentrates to achieve smelter limits. Arsenic can be removed from copper concentrates by the heat-induced decomposition of the predominant enargite mineral (Cu3AsS4) at ~700°C. Roasting at such elevated temperature is a commercially proven option, but still suffers from process, cost and environmental challenges. After roasting, the released arsenic sulfide fumes can be separated from most of the entrained concentrate dust in cyclones. Then, the As2S3 laden off-gas can be oxidized to generate a still volatile oxide (As2O3), which can be further separated from the residual concentrate dust in a hot electrostatic separator. Next, the off-gas can be cooled with air or water, to either deposit the arsenic oxide in a dry or dissolved state. Dissolved arsenic acid can be oxidized to As(V) and reacted with iron to yield a stabilized ferrihydrite precipitate or scorodite that can be safely disposed. In the case of dry arsenic trioxide, it can be mixed with reactants (e.g. iron) and fluxes before thermal stabilization in a glass matrix, thus eliminating the need for costly oxidation.

Jan 1, 2019

By F. O. Verret

Several grindability tests were developed over the years to design grinding circuits or optimize existing operations. Each test has its own strengths and weaknesses and it is imperative to select the proper test procedure(s) to meet project deliverables and minimize the risk of a project. Pilot plant testing of large bulk samples historically constituted the traditional approach for AG or SAG design, but was gradually replaced by small-scale tests. Nowadays, conducting grindability tests requiring only a few kilos of material on several samples is a more typical approach to grinding mill design. This paper summarises the requirements and deliverables of various bench-scale test procedures, their strengths and weaknesses, and cases where AG/SAG pilot testing should still be performed.

Jan 1, 2012

By John Kazakoff, Andrew Mortimore

"Production at the remote Kumtor Gold Mine in Kyrgyzstan commenced on schedule in December 1996.The ore mined during the first few months of 1997 from the upper benches of the pit did not produce the results predicted by the feasibility study. Low metallurgical recoveries experienced at the time were cause for great concern.It soon became clear the recovery difficulties centred around the preg robbing nature of the ore.Procedures were developed to selectively stockpile highly carbonaceous low recovery ore types. Testing methods were developed in the mill to quante the preg robbing activity of the ore. Preg robbing masking agents were added to the process to reduce the preg robbing activity of the ore.The importance of coconut carbon activity relative to the preg robbing activity of the ore was realized. Every aspect of the performance of coconut carbon in circuit was evaluated and optimized to improve carbon activity.Mineralogical analyses were completed defining the deportment of gold in ore and process tailingsBy year end the recovery difficulties had been overcome and plant design production had been achieved. Process improvements continue through 1998."

Jan 1, 1999

By Euler De Souza

This paper presents a study conducted to assess sillmat behaviour during undercut mining and to provide a comprehensive understanding of the effects of arching and sillmat thickness on stability. The design study investigated the effects of such parameters as stope width and height, closure strains, orebody geometry and inclination, wall roughness, fill properties and cure time on the stability of cemented sillmat during undercut mining. Arching effects have been quantified based on centrifuge modeling studies for the different mining conditions. Observations have indicated that the degree of arching is a function of stope geometry, wall roughness, fill cohesion, and wall closure. Arching effects increased with decreasing stope width, increasing backfill cohesion and frictional resistance, increasing wall roughness and with wall closure.

May 1, 2002

By Constantin D. Pitis

Similarly to notions of constant capital/expenses in business or magnetization power for electric motors an energetic base load (E0) is always necessary to set up a mining process. Denominated as ?non-productive energy? or base load, (E0) is a constant component of the entire energy consumption of the process, while another component the ?productive energy? (Ep) is variable, being function of the production/throughput (M). By using these new concepts, the author creates an energetic model proposing a new method of assessing changes in mining processes (throughput) that can reveal possible energy savings (DSM component) or energy losses. Proposed mathematical model of assessing (E0) and energy savings by using ?Marginal Energy Usage Index? (MEUI) enables creation of historical benchmarks and direct on line energetic (efficiency) assessment of mining processes and/or equipments. The paper analyses the case of assessing possible energy savings that can be obtained by: ?? Improving efficiencies of various equipments (by retrofit) ? reducing E0 ?? Improving the process control or other process changes ? reducing Ep ?? Monitoring persistence of the adopted Energy Conservation Measures (ECMs). For the new plant design, the proposed method enables prediction of Customer Baseline in various mine/process configurations. The method could become part of a ?preventive action? in design process by offering a reliable tool of assessing additional financial effort (Incremental cost) when choosing various higher efficiency equipments and/or processes.

Oct 1, 2009

By J. E. Nesset, L. S. Urbanoski, J. R. Goode

"The benefits and advantages of pre-competitive, collaborative industry/university research are not always well appreciated. This paper presents the case study of a current collaborative project between CAMIRO-Metallurgical Processing Division and McGill University that is investigating the understanding and prevention of sphalerite activation in lead and copper-lead flotation circuits. The benefits of the research findings to the sponsors can be maximized by an approach which links the researchers themselves with the sponsors at frequent, interactive project review sessions. Some examples of the findings and their application to industry are also presented along with a general update on CAMIRO-MPD.INTRODUCTIONThe history of Canadian efforts in the area of collaborative pre-competitive research is not long when compared to Australia (AMIRA) and Britain (MIRO). AMIRA (Australian Mineral Industry Research Association) began over 40 years ago and has had a well recognized role in bringing together industrial sponsors and researchers to fund a large variety of very significant development projects in that country, The total AMIRA funding effort for mineral processing work currently exceeds some $Aus 40 million with new projects valued at about $Aus 12 million initiated each year. This high level of commitment by industry has contributed to establishing Australian researchers and their facilities as among the best in the world.Canadian companies have traditionally funded much of their R & D in-house with extensive facilities and large funding commitments. The past 2 decades have seen most of these facilities disappear. The current situation is one in which only the largest integrated metal and mining companies have retained their own R & D groups (e.g. INCO, Falconbridge, Noranda). Industry-funded research at universities was primarily through individual efforts by the mining companies and rarely included collaboration with other mining companies, equipment or chemical suppliers, or engineering firms."

Jan 1, 2001

By O. Matwijenko

"1. IntroductionThe Falconbridge Owl Creek open pit gold mine is located 17 km northeast of Timmins and 2.5 km west of the Kidd Creek metallurgical site. Diamond drilling first commenced in 1964 with open pit production commencing in 1982. Other operations within close proximity are the Canamax-Pamour Bell Creek Mine and the Falconbridge Hoyle Pond Mine.Since 1982, 1,216,000 tonnes assaying 4.5 g/t have been processed in six concentrators. This paper summarizes the metallurgy achieved with special emphasis on some of the unique problems encountered in the processing of this refractory ore.2. GeologyThe Owl Creek deposit, located 3 km north of the Porcupine-Destor fault is in easterly trending carbonaceously altered volcanic and sedimentary rocks. The deposit strikes east and dips steeply north.The ore body can be best described as a wedge shaped sandwich of volcanic basalt between the surrounding sedimentary greywacke. The north and south volcanic-sedimentary interfaces are marked by graphitic zones. The south graphitic contact is more continuous with its thickness varying from 5 m to 20 m as compared to 0.5 m to 3 m to the north. Quartz veins occur throughout the deposit with a predominant vein running along the south contact where it is characterized by numerous graphitic inclusions. Some veins carrying gold values intrude into the sediments in the north. With depth the sedimentary and volcanic zones taper while the quartz-graphitic zones appear to become more dominant."

Jan 1, 1988

By K. P. Wright

"Twenty-five years ago, there were about thirty-two gold mines operating in Canada; nine of these have been in continuous operation since then and are still operating. Three of these are part of the Dome Group. Operations of these three and of the newer mines in the Group are referred to in discussing process development over the last twenty-five years.Of the Dome Mines Group, Dome Mines Limited, Sigma Mines (Quebec) Limited, Campbell Red Lake Mines Limited, Detour Lake Mine and Kiena Gold Mines Limited, three have been in continuous operation for the last twenty-five years. Dome has operated for seventy-seven years; Sigma about to complete fifty years, and Campbell approaching forty. Overall, milling processes are virtually the same as they were twenty-five years ago. Tota1 go1d production from the three mines has hardly changed; 1960 production, 416,700 ozs. versus 411,300 in 1985. Milling costs have risen by a factor of six or seven and gold price from $35/oz. in 1960 to $322 in 1985.Basic flowsheets are relatively unchanged. Sigma is an all cyanidation mill grinding in cyanide solution; Dome operates gravity concentration and cyanidation; Campbell gravity concentration, flotation, roasting and cyanidation. All three use Merrill-Crowe zinc precipitation.Detour and Kiena are two examples of the major change in gold processing in the last twenty-five years - carbon-in-pulp. This process has brought along with it other changes, such as, carbon stripping and reactivation, electrowinning, and induction furnace melting. Modern mill design with its use of outside thickeners and agitators, bulk handling of reagents, instrumentation, etc., has made a radical change in operating conditions and appearance of the new plants."

Jan 1, 1987

By Keith R. Vining

Solid fuel, usually coke or anthracite, is commonly added to ground iron ore fines in the pelletising process to reduce the amount of burner fuel required and the maximum firing temperature, while maintaining pellet strength and the productivity of the pelletising plant. However, the use of coke and coal contribute to the production of greenhouse gases (eg carbon dioxide) and are a source of NOx and SOx emissions. Utilising wood biomass / wood char is one methodology CSIRO Minerals is investigating as an alternative to coke and coal, as it is a renewable energy source that is very reactive, has a low ash content and the potential to reduce emissions of greenhouse gases from the iron and steel-making process. Previous work has established that charcoal can effectively replace coke in the sintering process (Lovel et al, 2005), and the following experimental work aims to demonstrate that it can also provide an adequate replacement for coke and coal in the pelletising process. The paper presents the results of an experimental testwork program comparing the behaviour of fired pellets containing traditional solid fuels and charcoal. The physical properties of the pellets are compared in terms of their compressive strength, tumble and abrasion indices and the reduction properties. A qualitative assessment of the pellet microstructures is also presented.

May 1, 2007

By M. Doggett

All 231 Canadian primary gold mines that produced from 1946 to 2004 were analyzed with respect to their production and economic characteristics. Trend analysis revealed that the average annual capacity of Canadian gold mines, as measured by ore processed or gold produced, has increased over time. Conversely, the average grade of gold mines has decreased over time. The distribution of gold output is highly uneven: for example, the 22 giant mines producing more than 100 t (metric tonnes) of gold account for 62% of total output. Small mines producing less than 10 t Au account for 54% of total mines, but less than 4% of total gold. Each historical mine in the database was evaluated on the basis of a set of economic and technical assumptions to determine if it would be economic to develop today. Based on the net present value using an 8% discount rate, only 103 of the 231 mines were determined to be economic. About 80% of the uneconomic mines had less than 10 t of contained gold, and 63% had less than 5 t. It is concluded that these small mines would not be viable today due to a combination of real capital cost increases, and more stringent permitting and environmental regulations. The implication is that future gold supply in Canada will hinge on the discovery and development of a new generation of larger mines. Given that only one giant mine has been discovered in the past 20 years, it seems likely that gold output will decrease in the future.

Jan 1, 2007

By Chuck Edwards

"Cameco operates the high-grade McArthur River uranium mine in northern Saskatchewan, and is constructing the high-grade Cigar Lake uranium mine. Cameco has studied the concept of underground milling of high-grade uranium ore. Compared to conventional milling on surface, the underground milling scheme appears to offer significant cost savings and a lower environmental impact. This paper describes the underground milling scheme, presents process flowsheets and plant layouts, and provides an assessment of potential benefits from underground milling.INTRODUCTIONCameco CorporationCameco is the world’s largest uranium producer, with three operations in Saskatchewan (Rabbit Lake mine/mill, McArthur River mine, and Key Lake mill), two in the United States (Crow Butte and Smith/Highland ISL operations) and two new mines being developed in Canada (Cigar Lake project) and Kazakhstan (Inkai project). Cameco provides uranium fuel refining and conversion services at its Blind River and Port Hope facilities in Ontario. Cameco is part owner (31.6%) of Bruce Power that leases eight nuclear power plants located in southern Ontario. Cameco is the majority shareholder in Centerra Gold Inc., which operates gold mines in the Kyrgyz Republic and in Mongolia.Global Uranium Production RequirementsCentral to Cameco’s view of future uranium supply requirements is the belief that the world is experiencing a nuclear renaissance. This belief is supported by a number of developments, including:•,?Countries representing half the world’s population are now constructing new nuclear power plants. Many countries without nuclear power have begun to plan for it, seeking reliability, clean air and energy independence.•,?Accelerated growth for this industry is also becoming much more likely. In Asia, primarily in China and India, 92 reactors are under construction, on order or planned to help meet rapidly expanding economic needs. Other Asian countries, such as Vietnam and Indonesia, with large populations and rapidly growing economies, are also expected to follow.•,?In the United States the “Energy Policy Act of 2005”, the first energy legislation introduced in over a decade, was signed. This legislation includes provisions recognizing, for the first time at a national level, the clean air contribution of nuclear energy and its critical role in meeting energy security. Upon signing this landmark legislation President Bush stated, “We will start building nuclear power plants by the end of this decade.”"

Jan 1, 2007

By Glenn S. Dobby, Glenn A. Kosick

"Column flotation circuit engineering has matured to a good level of understanding regarding testing, scale-up and design. Some of the factors important to the process of implementing columns are addressed in this paper, highlighting the following issues• operating bias and the maximum achievable solids carrying rate for a column application.• internal circulating loads of full scale and pilot plant flotation columns• short-circuiting oflarge scale columns and the impact on separation selectivity• the number of stages in a column circuit and the impact on selectivity.The paper uses several case studies, from pilot plant to full scale implementation, to illustrate these design issues.1. IntroductionTen years ago, in 1984, there were very few column flotation installations world wide and the largest columns were about I . 5 to 2 m diameter or square. During the following decade there have been in the heighbourhood of 80 column installations that have been successful and profitable, as both retrofits and green-fields plants. Now, columns as large as 3 to 4 m are not uncommon. Examples of installations with many large columns include Escondida, with ten columns each at 4 m square, and P T Freeport Indonesia, with 21 columns: nine 3 m diameter columns, four 2.4 m diameter columns and, starting up in April I 995, eight 3.6 m diameter columns.During this growth period there have also been column installations that can be considered as failures. In the majority of these cases, the reason for their failure has been an insufficient installed capacity for applications where there is a high weight recovery of solids to the froth product These include zinc cleaning, copper cleaning and iron ore flotation (silica flotation from a feed that is 15 to 30 %Si02) An application with high weight recovery to the froth product makes column scale-up and circuit design more complicated. This is particularly so if a high degree of selectivity is also required while attaining a high weight recovery, which can happen with iron ore flotation in columns."

Jan 1, 1995