Search Documents

Building off the enthusiasm generated by Credit Suisse First Boston on the potential benefits of an industry-wide effort to market gold jewellery, a group of gold producers, led by Randall Oliphant, CEO of Barrick Gold, decides to investigate the long -term impact of launching a significant marketing initiative to stimulate demand for gold jewellery. The group establishes an international Steering Committee representing producers of differing sizes, members and non-members of the World Gold Council. This Committee comprises representatives from AngloGold, Barrick, Gympie Gold, Kinross, Randgold, and Sons of Gwalia and is supported by the Australian Gold Council, the Gold and Silver Institute, and CSFB. The Steering Committee hires McKinsey to assist it by conducting an independent review of whether investment in a large-scale marketing initiative for gold jewellery would be effective in increasing demand for gold, and ultimately of significant economic benefit of the gold industry. Based on McKinsey?s review* and its own discussions with key stakeholders across the industry, the Steering Committee strongly believes the case for marketing is compelling and sufficiently proven to persuade the industry to undertake a major initiative over the next few years. The Steering Committee, supported by McKinsey, conducts briefings with a range of gold producers to share its findings and to understand producers? perspectives on an industry-funded marketing campaign. The Steering Committee is asking for your support to bring this industry-funded marketing campaign to life.

May 1, 2002

By M. T. Ho, D. E. Wiley

The iron and steel industry accounts for almost 5 % of the total world CO2 emissions. Optimisation of the steelmaking process has led to significant reductions in the energy intensity to the point where it is now close to the thermodynamic limit. Further reductions are not likely to be achieved with efficiency improvement alone and will require the development of novel or breakthrough technologies. This paper will show the costs of implementing CO2 capture at direct atmospheric emission points for a number of iron and steel production processes. These include a conventional process, top gas recycling processes, the Corex and Hismelt smelt reduction processes and the Midrex direct reduction process. Depending on the point source emission, costs of capture with existing chemical absorption technology at iron and steel plants can be similar to capturing CO2 at pulverised coal fired power plants.

Jan 1, 2015

By T. Salman

Most authorities agree that the dissolution of gold from an ore takes place according to the Elsner equation first stated in 1846. The main requirements for the dissolution reaction to proceed are:(1) that the gold be clean.(2) that the gold be exposed to the solution.(3) that the cyanide solution be active, i.e., not over loaded with impurities which could retard or inhibit the reaction.(4) that an adequate supply of oxygen be available (from air).When the above conditions are met, cyanidation becomes basically a mechanical operation.Minerals such as quartz, calcium carbonate are relatively inert to cyanide though they may give rise to mechanical troubles. Most ores present an assortment of chemical problems, depending upon the constituent present, which react to the solution to varying degrees. It should be borne in mind that minerals which are disturbing to the operation, though present in small amounts, usually exceed the weight of the gold by several hundred times. In addition, the application of flotation and other forms of concentration to gold ores, previous to cyanidation, concentrates not only the gold, but also the other heavy minerals and these complicate the chemistry of the process.

Jan 1, 1975

By E. W. Williams

THE Ontario Research Foundation has developed a process for producing low-density steel shapes by direct reduction of iron oxide powders. The process is a combination of reduction, sintering (without compaction), and carburizing, and has been described by Cavanagh (1). Iron oxide powder, in the form of mill scale or highly concentrated iron ore, carefully sized, is poured into a porous mould. A reducing agent, generally coke breeze, is placed around this mould in a heat resistant sagger which is sealed off and heated to 2,000°F. for a number of hours. The coke/ ore ratio, the particle size distribution, the temperature and time of soaking, and the size and shape of the reduced product, determine the densities and carbon contents attained. The Defence Research Board sponsored a programme of research into some of the mechanical properties of the material (which was named Controlled Density Steel) and this paper is an account of various factors which govern the tensile strength of the steel. The ?practically attainable range of density is from 1.5 to 5.5 gm./ cc. but higher densities are obtained on hot working. This paper, however, is restricted to discussion of results for moulded products.

Jan 1, 1957

By A. Gakwaya, K. Ba, M. Guillot, J. Levesque

"In the aerospace industry, parts requiring tight dimensional tolerances and high mechanical properties are often produced by forging. This forming process allows for an optimal use of the material and can produce high quality parts at a reasonable cost. Thanks to CAD techniques, it is now easier to predict flow profiles and in-process defects occurrence as well as wear of the forging dies. However, in order to determine the properties of the forged part with computer simulation methods, a good material constitutive law is a prerequisite. This paper presents a virtual manufacturing process modeling for the closed die forging of a workpiece performed with Abaqus. The hot closed die forging is a regular process applied to manufacture metallic components used in aerospace industry. In this forming process, the material flow and final properties depend on the microstructure evolution during the process that is why it is important to include the microstructure effects in the material model. Based on the stress-strain data measured experimentally, the material parameters of the standard (classical) Johnson-Cook constitutive model and of a modified Johnson-Cook constitutive law including the microstructure effects have been determined. The obtained simulation results are compared with the experimental results for validation.INTRODUCTIONDue to their good physical and mechanical properties such as ductility, fatigue strength, corrosion resistance and rupture, aluminium alloys are often used in the aeronautic and the automobile industries. In this work, we present the forging of an Al-Zn-Mg-Cu alloy, a high strength aerospace alloy recognized for his good combination of physical and mechanical properties. The forging process is simulated in Abaqus Explicit using both the standard Johnson-Cook law (Johnson & Cook, 1983; Hor, 2011; Lin et al., 2012), and its modified version as proposed by Andrade et al., (1994). The Johnson-Cook law offers the advantage of considering the coupled effect of the plastic strain, the plastic strain rate, the temperature of forging and the hardening effect. However, it is a phenomenological model which does not account for the effect of the microstructure evolution. In this work, a modified Johnson-Cook model implemented in a user subroutine is used. It includes the effect of the dynamic recrystallization phenomenon during the hot deformation. The dynamic recrystallization has a noticeable influence on the microstructure and on the properties of the material. The term dynamic means that the process starts and acts during the deformation and this is different from the static mechanism of recrystallization which occurs during an annealing and after cold or hot deformation. Since properties of the alloy used in this work are not readily available in the literature then characterization tests have been done at various temperatures and strain rates, in order to determine the thermomechanical properties and the material behavior during the process of hot forming, more precisely during the hot forging process."

Jan 1, 2016

By L. McFarlane, D. Rollwagen

"Madawaska Mines produces a uranium precipitate (83-85% u3o8) using a magnesium oxide slurry to precipitate the uranium from solution. The solution, at approximately 9""0 g/l u3o8 is neutralized to a pH of 7.0. At the Eldorado Nuclear Refinery in Port Hope, the precipitate is not acceptable as a single feed to the refinery process. It was discovered that a 2% silica impurity caused emulsion problems in the solvent extraction column.After extensive studies of possible solutions, such as two-stage precipitation and solvent extraction, an acceptable solution was found. The selective precipitation of uranium at a pH of 3.5-4.0 using Hydrogen Peroxide produces an almost silica free product of higher purity (90-93% u3o8) to that of straight magnesium oxide precipitation. Introduction Madawaska Mines Limited is located four miles southwest of Bancroft, Ontario. The annual production of u3o8 is approximately 610,000 pounds. To produce the final product or yellowcake, nine separate steps are required. These are (1) grinding, (2) dewatering, (3) leaching, (4) acid filtration, (5) clarification, (6) ion exchange, (7) precipitation, (8) washing, and (9) drying and packaging. To illustrate these steps, a complete flowsheet of the mill is displayed on page three.The yellowcake is sent to Eldorado Nuclear Limited in Port Hope, Ontario for further concentrating and refining. As part of their routine specification check, a test called the amenability test is performed. This test indicates how a particular dissolved yellowcake product reacts to solvent extraction. The Madawaska yellowcake did not pass the test."

Jan 1, 1982

By J. Duncan Keppie

"Auriferous quartz veins occupy zones of dilation. Foldgenerated dilation zones are systematically distributed and are dependent upon the folding mechanism. Comparison of natural and theoretical folds indicates that different mechanisms operated in the Halifax and Goldenville formations.In the slates of the Halifax Formation, the folds are of similar type and internal strain is most intense in the limbs. This internal strain is probably accommodated with little or no dilation.In the meta-sandstone and slates of the Goldenville Formation, a combined flexural and tangential longitudinal strain mechanism, modified by pure strain, produced the folds. Tension fissures produced by this mechanism are: (a) saddle 1'eef veins in slate horizons, (b) veins normal to bedding in meta-sandstone layers in the fold hinges,(c) veins along faults in the cores of tight fold hinges and (d) zones of en- échelon veins in slate honzons in the limbs of folds. Because veins in fold hinges (a, b and c above) account for most of the gold production in the Meguma Group, other factors controlling the distribution and size of these veins are summarized below.1. The size of saddle reef veins is controlled by the curvature of the fold hinge. Thus, a Curvature Index greater than 600 degrees indicates a promising prospect, where"

Jan 1, 1976

By S. Masambi, S. M. Bradshaw, C. Dorfling

The presence of iron in base metal leach solutions complicates nickel and copper recovery processes. Leach solutions are commonly purified by precipitating iron as hematite, jarosite or goethite. Fast oxidation of ferrous to ferric is an important step to remove iron by precipitation at low pH values. This study reviews the ferrous oxidation mechanism and reaction rates in chloride media using molecular oxygen. This work contributes to the development of a better understanding of iron precipitation processing steps, which continues to be an important matter in hydrometallurgy. The specific objectives were to investigate (1) the effect of key process variables (acid concentration, copper concentration, and temperature) on the rate and extent of ferrous oxidation by dissolved oxygen and (2) the oxidation mechanism by varying chloride and hydrogen ion concentrations separately. Batch oxidation tests were performed using synthetic chloride solutions containing 45 g/L Fe and 3 g/L Ni at 50 and 80°C, dissolved Cu concentrations (0.3 to 3 g/L) and HCl concentrations (2 to 4.6 M). Regression analysis indicated that the significance of the effect of different variables on the extent of oxidation achieved after five hours decreased in the following order: HCl concentration > Cu concentration > temperature. Increasing the HCl concentration from 4 M to 4.6 M caused the degree of oxidation achieved after two hours to increase from 58.3% to 98.1%. The catalytic effect of dissolved copper on ferrous oxidation was more significant at low acid concentrations. Experimental data suggest that rapid oxidation of ferrous is possible at high HCl concentrations for two main reasons. First, the formation of ferrous chloride complexes allows ferrous oxidation to proceed without the formation of superoxide as an intermediate product (the rate limiting step in dilute acid solutions). Second, the formation of ferric chloride complexes stabilises the oxidation product.

Jan 1, 2016

By Jean-François Dorion

In an underground mine rock reinforcement elements and surface support units are often exposed to conditions that make them susceptible to corrosion. These unfavourable conditions include high relative humidity, the presence of dust and various other particles in the atmosphere, splashing of water and gases from blasting and diesel equipment. The corrosion of support systems is often associated with a loss in capacity that can be a major safety and economic concern in underground mines. A support system is composed of several components such as bolts, mesh, plate, retaining nut, cable ties. Every individual component can deteriorate due to atmospheric or aqueous corrosion and consequently can damage the integrity of the support system. This paper reports the results of a 5-yr study on the loss in capacity of individual support and reinforcement components at selected underground hard rock mines. As part of this investigation corrosion coupons were installed at selected locations at participating mines. A number of these coupons were recovered at regular time intervals and analysed for any signs of corrosion and subsequently tested to quantify any loss in capacity. The testing of the corrosion coupons was complemented by similar tests on recovered mesh and friction bolt specimens from the site. The collected data have provided useful insights on the factors that contribute to the corrosion of support systems. This knowledge can aid in the selection of appropriate support strategies and a reliable assessment of the predicted useful life of a support system and the need for rehabilitation.

Jun 1, 2012

By C. A. Campbell

"There has been a significant reduction in maintenance hours and improved performance by maintenance crews at the MacLeod mine since the introduction of a Maintenance Control system which includes the use of maintenance standards. This paper deals with the steps taken to develop and implement the program. The objectives of the program and the organization and training of the control group are outlined. The work-order system, standards application, classification of work and the preventive maintenance procedures will be discussed.All maintenance information is processed on a Model 360-20 IBM computer, which turns out maintenance records and control reports for management action. Information in these reports, and reasons for reporting, will be discussed and the benefits resulting from the program reviewed."

Jan 1, 1973

By F. Andrieux

"Extruded aluminum profiles are increasingly used for lightweight vehicle construction. Since fracture strains of aluminum profiles are relatively low, damage modeling is crucial for reliable crash simulation. For aluminum profiles, not only the stress state but also the orientation influenced both the deformation and damage behavior. For the material characterization smooth tensile tests were performed in six orientations, both stress-strain curves and r-values were registered. Moreover notched tensile and shear-tension tests were performed in three directions. Finally punch specimens were tested to investigate biaxial loading. Digital image correlation (DIC) analyses were performed to determine local strain values such as critical values at failure. Numerical investigations were conducted with the strain based damage model GISSMO, which is well suited and widely applied for crash simulation. For a strain based failure model the choice of the deformation model is crucial. Deformation models with increasing complexity were investigated to find the best compromise over all experiments. It was found that there is still a great requirement on development of a material model which describes the orientation dependence of the hardening behavior over all stress states in order to be able to use a strain based failure model reliably.INTRODUCTION Weight reduction is an important step for less energy consumption and as a consequence aluminium is increasingly used in automotive parts. An important field of application is the well-established aluminium extrusion which allows the realisation of quite complex shapes of profiles required for an innovative lightweight design with integrated functions. Typical applications are chassis parts, bumpers, crash elements, air bags, etc. Especially the Al–Mg–Si medium strength alloys of the 6000 series can be easily extruded to form complex profiles and prevail for extrusion. Since fracture strains of aluminum profiles are relatively low, damage modeling is crucial for reliable crash simulation. For aluminum profiles not only the influence of the stress state but also the orientation with respect to the extrusion direction have an influence on both the deformation and damage behavior and need to be characterized. The effect of orientation on deformation behavior has been intensively investigated in the last decades and material models haves been proposed (e.g. Barlat & Lian, 1989; Barlat, Lege, & Brem, 1991; Barlat et al., 2003) to describe accurately anisotropic behavior. However it can be noticed that the parameters of anisotropic material models often rely on the yield stress in different directions, which means that the effect of strain hardening is not automatically accounted for. The accuracy of the models at large deformation is not guaranteed and has to be verified. In contrast to deformation models both the characterisation and the modeling of anisotropic damage behavior are not systematically studied. Isotropic failure models based on a critical strain as function of stress triaxiality (ratio of the hydrostatic stress to von Mises equivalent stress) are widely used for crash application (e.g., Bai & Wierzbicki, 2010; Sun, Andrieux & Feucht, 2009; Sun et al., 2013). The mechanisms of failure are manifold and not fully understood; to develop an anisotropic failure model it is mandatory to also have experimental data in different orientations for loading scenarios other than uniaxial tension, e. g. shear, biaxial tension, which implies an extensive experiment plan. This makes the development of failure models more challenging than deformation models for which it is possible to develop a reasonable anisotropic model using the framework of yield function and calibrate the material parameters based on results of tensile tests in different directions. Moreover the development of deformation and failure models are intricate as the variables governing the damage evolution are predicted by the def"

Jan 1, 2018

By N. E. Izatt

Molecular recognition technology (MRT) has a wide range of applications in the hydrometallurgical and chemical processing industries for the selective removal of iron as well as various other impurities from a variety of streams, including aqueous solutions through a wide pH range, as well as organic solutions. An example of a commercial application is provided for the removal of iron from an aluminum chlorohydrate stream. In this example, MRT has a selectivity for iron relative to aluminum which is significantly higher than that of ion exchange resins. A discussion of MRT capabilities for other impurities and streams is also provided.

Jan 1, 2006

By M. Pierce

Empirical relations exist for the estimation of rock block strength at scales larger than standard core sizes. Two recently published relations account for the significant impact that microflaws (pores, open cracks, veins) and weathering/alteration can have on scale effect. These relations were used to estimate the rock block strength of heavily microfractured quartzite from the Bingham Canyon Mine. Predictions were compared to the results of Uniaxial Compressive Strength (UCS) tests conducted on cylindrical core samples of varying microfracture intensity and diameter. The comparison confirms the sensitivity to microfracturing suggested by existing relations and suggests that they would benefit from more detailed consideration of microfracture strength, density and persistence over a range of scales. The results of a numerical study illustrate how Synthetic Rock Mass (SRM) modeling techniques could be used to complement empirical data in the development of such relations.

May 1, 2009

By R. D. H. Wigmore

"The title of this paper probably sounds quite simple and self-explanatory. The problem, however, was not a simple one, and proved somewhat baffling and expensive over a period of years. After outlining the problem, and the methods which were used to overcome it, we will present our solution, or perhaps it would be more accurate to say the partial solution which we reached.We have in the cyanide circuit at Upper Canada two Dorr Single Tray Thickeners with 30 ft. diameter wood stave tanks 12 ft 4 1/ 2 in. deep. In the years prior to 1948, when the tonnage was 250 to 280 tons per day, two stage thickening was practised. As the tonnage was increased to 450, 500, and finally 560-600 the thickeners were used in parallel.Some settling problems were encountered, and in those days the means to combat these problems were very few. The only remedy was to dump in pounds and pounds of lime. Then along came a few reagents, and the writer remembers one in particular, - 11 Carboxel11 (carboxymethlycellulose). When water or solution was added to this reagent, the result was a gooey mass somewhat resembling tapioca. One drawback was that it stayed in this state, and didn't go anywhere. A method was finally developed of containing the mass (or mess) on a fine screen, and dissolving some of it was a fine spray of solution. The only thing the writer recalls was that it was not a very effective settling agent.Shortly after this time, along came the new flocculents, first a few, and then many, and our settling problems were over.But when the settling problem was solved at Upper Canada, some other problems were creeping up on us."

Jan 1, 1965

By Henri F. Letient

At the 25th Annual B.C. Mine Reclamation Symposium in September 2001, Huckleberry Mines Limited received the Jake MacDonald Reclamation Award for its exemplary mine waste management plan, conceived to prevent the onset of acid mine drainage and facilitate reclamation of the site. The mine is an excellent example of a total mine waste management plan, with all materials scheduled in detail for the life of the mine. The filling plan for the waste management facility takes into consideration the requirement for permanently flooding the potentially acid generating rock. From an acid generation point of view, the tailings fall in the zone of potentially acid generating material. To avoid long-term ARD concerns, it is planned to float pyrite out of the tailings towards the end of the mine life, thereby generating a non-acid generating tailings stream which will be used to cap the deposited mine waste. The pyrite-rich tailings will be discharged into the deeper portion of the impoundment. On closure, a permanent water cover will be maintained over the mine waste. Only beaches of non-acid generating tailings will remain exposed. Mining will be phased to also allow one of the pits to provide storage for potentially acid generating rock and to be flooded at the end of mine life.

May 1, 2002

By W. R. Richmond, T. S. Li, D. Ilievski

The reactive precipitation of ferrihydrite was studied in a laboratory semi-batch precipitator, where a highly concentrated sodium hydroxide solution was fed into an iron nitrate solution. Instantaneous hydrolysis of iron[III] takes place when the two reagents are mixed generating supersaturation, and leading to immediate precipitation of ferrihydrite via a primary nucleation mechanism then followed by limited crystallite growth to 3 to 7nm. A mathematical model of the semi-batch laboratory precipitator was used to simulate the pH responses during the precipitation of ferrihydrite at different reagent feed rates. The model comprised of interconnected compartments of averaged hydrodynamic conditions, incorporating physically meaningful turbulent mixing time constants. The precipitation process is described by the solute concentration conservation equation incorporating a postulated ferrihydrite primary nucleation model. It was found that the semi-batch precipitator model predicted the experimental pH response reasonably well. The experimental results demonstrated that the reagent feed rate has a critical effect on the course of the precipitation of ferrihydrite, i.e. it is a rate determining step given the fast hydrolysis and precipitation kinetics.

Jan 1, 2005

By F. Mackay, S. A. B. da Fontoura

"The theoretical solution for a viscoelastic well was presented by Gnirk and Johnson in 1964. From this an exercise was executed to validate the numerical simulation of a cemented salt well. The exercise comprised the geomechanical behavior through time of a cemented salt well, this was achieved using a finite element commercial software. This simulation showed the accuracy of the theoretical solution for viscoelastic materials such as salt rock. The theoretical solution presented elastic and viscoelastic analytic equations for the tangential and radial stresses. The equations for the elastic material (casing and cement) and for the viscoelastic material (salt rock) are presented here: Where equations 1a & 1b are the radial and tangential stresses of the elastic material; equations 2a & 2b are the radial and tangential stresses of the viscoelastic material; and a, r and b are the geometrical parameters explained in Figure 1. These equations are function of Pb, a stress time dependent factor that comes from the Burgers model of viscoelasticity. This is well explained in Gnirk and Johnson (1964) and for the sake of simplicity is not disclosed here. Po is the stress at section A-A’ dependent of the depth h."

Jan 1, 2015

By Eric Toh, Michel David

"Geo-statistics was born from mining problems: grades not being what was expected; more tons of waste than expected; in other words, dilution. Geo-statisticians now claim to have solved many of these problems, but the solutions seem to have a hard time being picked up by practitionners. One of the reasons may be the jargon which geo-statisticians are using. The reason for jargon is that obtaining solutions required a strict formulation in a statistical framework. Practicing mining engineers and geologists are not used to such a framework, and they are not familiar with the jargon. In this paper, the authors re-examine dilution in operating mines and answer two extremely practical problems, using practical language as much as possible. Can dilution be reduced by improving the quality of blast hole samples? How many samples are needed to obtain a good ore-waste discrimination, i.e. to reduce dilution to acceptable levels? Citing real situations, this paper will try to show how fairly simple concepts can be used to answer both problems. It will try to present the solut ions using a minimum of jargon, even if the theoretical solution could be obtained in a few precisely worded paragraphs.The two problems considered are the following: the first one was encountered in the Panguna deposit of Bougainville Copper Limited, P.N .G., but it occurs in every open pit operation where selective mining is required. As the cut-off is lowered, more and more mining occurs in marginal areas. Grade control is mostly based on the grades obtained from blast holes, hence, it is essential that the quality of the blast hole samples be as good as possible in order to reduce dilution, waste being sent to the mill and ore being sent to the waste dump.The second example stemmed from the grade control technique used in many gold open-pit operations in Western Australia: channels are cut on lines 5 m apart and assays are taken every metre. The pattern in the same regardless of rock type, whether one is in the oxidized zone or not. The number of samples to be treated is enormous, hence, the question arises how to find an acceptable sampling grid, and whether changing it can reduce dilution at an acceptable cost.A solution will be given to each problem separately, although in fact it could be seen that the formal framework is the same in both situations."

Jan 1, 1989

By M. N. Vu, G. Armand, A. Noiret, D. Seyedi

Extensive experimental observations have been performed around drifts and shafts at the Meuse Haute-Marne (North-Eastern France) Underground Research Laboratory (URL) to assess the extent and pattern of the induced fracture networks in Callovo-Oxfordian claystone. Two types of fractures can be distinguished, namely; shear (mode II) and extension unloading (mode I) fractures. The main goal of the present paper is to provide insights on the extent of damaged zone around the drifts by numerical simulations. A two-step approach is considered. First, 2D elastic calculations are performed for simulating the excavation of a drift following the direction of the major in situ horizontal stress. The evolution of the stress and strain tensors around drifts are calculated by applying an elliptic convergence function on the drift wall. Two failure criteria are then verified for the induced damage. An equivalent strain and an elastic energy based damage criteria are considered. The obtained results show that the employed simplified approach provides rather good results for the extent of the damaged zone for drifts parallel to major horizontal stress. However, post-processing elastic results leads to overestimating the vertical extent of the damage zone and underestimating its lateral extent, where the in situ stress state is quasi-isotropic in a section of drift. Nevertheless, the obtained results show that an energy based damage criterion could be considered as a potential way for reproducing the extent of the damaged zones if a convergence function can be defined for the drift.

Jan 1, 2015

By G. Stokreef, B. R. Davis, M. Kadiyski, Alvear Montenegro, H-P. Loock, V. Montenegro

An optical probe for burner monitoring in copper flash smelting, first developed at the University of Concepcion, is being industrially tested in a collaboration between Kingston Process Metallurgy, Aurubis AG, and Queen’s University. The sensor uses fibre-optics and optical emission spectroscopy to analyze the light emitted from combusting particles in the furnace reaction shaft. Information that can be extracted from the emission spectrum includes the temperature and brightness of the flame, as well as emission lines or bands from excited atomic and/or molecular species. Laboratory proof of concept testing of the sensor was first completed in a continuously fed drop tube reactor using O2/S stoichiometric ratios between 0.8–4.0. These O2/S stoichiometries capture the industrial operating condition, which is typically less than 1. The results from lab testing justified a plant trial of the optical probe, which was done in the Aurubis Hamburg flash furnace in September 2018. From this trial, approximately 31 hours of spectral data were collected and synchronized with furnace operating data. The results from both drop tower test work and the plant trial are discussed with implications for process monitoring in the commercial furnace. INTRODUCTION Flash furnace smelting is a continuous feed, batch tap, pyrometallurgical process. It is used to treat copper sulphide concentrates and small amounts of secondary materials, such as dust generated by the process. O2 enriched air and the concentrate blend are fed into the furnace reaction shaft through the concentrate burner. As the particles fall through the reaction shaft, they heat up and ignite, causing rapid, exothermic reactions that oxidize the iron sulphide, some copper sulphide and impurities in the feed. Reacted particles fall into the settler of the furnace, melt and separate into a matte and slag, while the SO2 enriched off-gas is carried out of the furnace for heat recovery and H2SO4 production. Operation of the furnace is achieved by adjusting a limited number of control variables. The main variables are the blast feed rate, blast oxygen enrichment, SiO2 flux feed rate and CH4 combustion rate (CH4 is used when the chemical energy of the concentrate is low) all measured relative to a set feed throughput. (Davenport & Partelpoeg, 1987; Davenport, King, Schlesinger & Biswas, 2002) The batch tapping nature of flash furnaces makes operation difficult because these control variables impact each other, and product quality can only be measured when matte and slag are tapped from the furnace. This means that operation of a furnace is based on feedback control from matte and slag assays. Because of the time-delay associated with understanding product quality, there is interest in using technology to monitor the combustion reactions that happen in the furnace reaction shaft.

Jan 1, 2019