Search Documents

By C. Montoya

This paper shows, through a case study, the impact of multivariate grade modelling upon mine design and mine planning. A deposit explored by drill holes is considered, in which the grades of five elements (copper, silver, molybdenum, arsenic, and antimony) are of interest. Forty alternative models of the deposit are constructed by fitting the joint correlation structure of the grade variables and using conditional cosimulation. In addition, a reference model, obtained by averaging the alternative models, is also considered. The study shows that the resulting mine design (final pit characteristics and production schedules) is sensitive to the grade model under consideration, and that the design based on the reference model may not be optimal when compared to the alternative models based on cosimulation. However, when assuming a given long-term plan and extraction sequence, the grades and net present value (NPV) calculated on the reference model are unbiased with respect to those calculated on the alternative models with the same extraction sequence. The latter allow assessing the possible dispersion of the actual grades and NPV around their expected values, and are useful for the planner in order to determine the probability of meeting given production targets and of exceeding or falling short of given threshold grades. Additionally, unlike cosimulation, the separate simulation of each grade variable leads to unrealistic resource models and to biased results in mine design and mine planning. This approach should therefore be avoided, unless the grade variables are spatially uncorrelated.

Jan 1, 2012

By S. A. Hosseini, O. Asghari

"Recoverable mineral resources assessment has become a standard geostatistical application in the mining industry, with various geostatistical techniques currently available. This investigation aimed to improve the prediction at the Sungun deposit and to obtain mineral resource models capable of handling (i) the change of support from drill-hole composites to selective mining units (SMUs), (ii) the multivariate nature of the ore control selection criteria that involves both copper and molybdenum grades; and (iii) the local uncertainty on the true (unknown) grades. The solution presented in this paper is to use direct block-support sequential cosimulation in order to construct a set of alternative outcomes of the copper and molybdenum grades on block support over the deposit, with no need for storing point-support values, hence with a considerable gain in memory management and CPU time. The grade realizations so obtained are then processed to calculate the uncertainty in the mineral resources that can be recovered above given cut-off grades, both globally and on a blockby- block basis. IntroductionGeostatistical methods are widely used by mining companies to determine a resource model of the tonnage and head grade of a potential orebody, which is one of the first and most critical inputs underpinning any mining project. The prediction of the grade, tonnage, and recoverable metal for a particular mining plan and the corresponding financial forecast constitute the main technical risks in mineral resource evaluation. The conventional approach to this problem is to predict the mineral grade for constant volumes relevant to the mining plan and to base the recoverable mineral resource calculations on those predictions. Alternatively, one can directly predict the local variability in the mining block grades based on a spatial distribution model (David 1988; Rossi and Deutsch 2014). Multiple realizations can also be constructed by conditional simulation, providing a more complete representation of block grade uncertainty, as well as the uncertainty over multiple blocks. Simulation methods allow quantification of the uncertainty of the mineral resource prediction risks in downstream studies such as mine design, mine planning, or operational optimization studies; the risk assessment is achieved after applying transfer functions to the conditional simulation models (Dimitrakopoulos 2012; Rossi and Deutsch 2014)."

Jan 1, 2017

By K, J. MAUNDERS

This paper briefly describes a computerised model of the South African electricity system which has been developed jointly by the Atomic Energy Board and the Electricity Supply Commission. The results of an analysis of this system have been subjected to a fairly extensive sensitivity analysis which reveals that up to the year 1990 between 680 and 700 million tons of coal will be required for electric power generation, together with between 7 000 and 10 000 tons of uranium. This result is fairly insensitive to data variations, but the prediction of a cumulative requirement up to the year 2000, of between 1 300 and 1 800 million tons of coal and 25 000 to 70 000 tons of uranium, proves to be very data-sensitive.

Jan 1, 2014

By S. Banisi, A. Ghasemi, E. Nematollahi

Discrete element method (DEM) has been used as a popular simulation method to verify designs by visualising how material will flow through complex equipment geometries. Although DEM simulation is a powerful design tool, finding a DEM model which includes real material properties (e.g., shape, cohesion and friction coefficients) is not computationally feasible. In order to obtain more realistic results, particle energy loss due to rolling friction has been highlighted by many researchers. To implement a reverse torque to account for rolling friction, various models have been proposed. On account of the complexity of the problem, there is no unique model for all applications (i.e. dynamic and pseudo-static regimes). In this paper, an in-house developed DEM software (KMPCDEM ©) was used to assess the robustness of three commonly used models by comparing the repose angle of iron ore pellets obtained through the draw down test in a laboratory setup and then the elastic–plastic spring dashpot model was modified based on considering the individual parameters instead of relative parameters of two contact entities. Results showed that the modified model could produce higher repose angle. The modified model was used to calibrate DEM simulation of repose angle of iron ore pellets in a laboratory setup of the draw down test. Comparison of the calibrated DEM simulation and laboratory results showed 2° differences between the predicted and measured angle of repose. This difference was attributed to considering the particle shape effects by assuming a spherical shape for particles. Keywords: DEM, rolling friction,particle shape, repose angle

Jan 1, 2020

Bougainville Copper Limited (BCL) operates a 50 Mtpy copper/gold/silver mine and concentrator on Bougainville Island, Papua New Guinea. BCL has installed an in-house developed data logging and supervisory process control system using a DEC PDP 11/73 mini-computer in the mill and has recently installed a similar system in the crushing plant. Together with the central DEC VAX 11/780 computer, they form part of a high speed, fibre optic, local area network in the concentrator. The in-house developed software, known as BIAS (Bougainville Copper Information and Automation System), was developed as a low cost option to interface to a number of different existing plant input/output systems while providing the capability to make maximum use of computer networking facilities. The package includes the more usual control system colour displays and data logging features, but also makes provision for two novel features: (a) Any user connected to the computer network can access the system displays. This allows metallurgists and management to monitor (but not control) plant operation from their offices. (b) Specialized application programs for performing advisory and supervisory control, dynamic simulation, and data analysis are easily incorporated into the system. The mill process system currently runs rougher/scavenger flotation stabilizing control and general data logging. The crushing plant system currently performs data logging only; however it will soon be augmented with the facility to run dynamic crushing plant simulations. Future developments will concentrate on making greater use of the detailed information available from the mine's fully automatic truck dispatch system which runs on a VAX 11/780 computer. Milling process control will be expanded to examine possible economic control of the operation of regrind and cleaner flotation circuits. An advisory water management system to optimize water supply within powerhouse generating capacity constraint will also be investigated.

Jan 1, 1987

Mixed integer programming (MIP) has been used for optimizing production schedules of mines since the 1960s and is recognized a shaving significant potential for optimizing production scheduling problems for both surface and underground mining. The major problem in long-term production scheduling for underground orebodies generally relate to the large number of variables needed to formulate a MIP model, which makes it too complex to solve. As the number of variables in the model increase, solution times are known to increase at an exponential rate. In many instances the more extensive use of MIP models has been limited due to excessive solution times.

Jan 1, 2010

By Y. C. Kim

The estimation of local recoverable reserves using the widely spaced exploration drill hole information in an open-pit mine is probably one of the most difficult tasks in ore reserve estimation. The difficulty of this problem is magnified if the grade distribution is highly skewed, such as in a gold deposit. This paper describes the results of performance comparison of three different kriging techniques in an open-pit gold - silver deposit. The three different kriging techniques compared in their local recoverable reserve estimation capabilities are: (1) lognormal kriging, (2) indicator kriging, and (3) probability kriging. For performance comparison, the predictions are compared with the actual recoverable reserves determined from blast hole information. Based on the results from one test bench, both indicator and probability kriging performed better than lognormal kriging, perhaps owing to samples not being truly lognormal. Between indicator and probability kriging, probability kriging results were definitely superior to indicator kriging results.

Jan 1, 1987

The theory of geostatistics covers a branch of .applied statistics aimed at a mathematical description and analysis of geological observations. Geostatistics can be used in pure geology (for example, for the analysis of trace elements in a metamorphic rock), in mineral exploration (for example, for the analysis of geochemical exploration data), as well as in mine valuation. This book is intended to provide a practical introduction to the theory of geostatistical methods of mineral evaluation. Over the years, various mathematical models have been developed to represent the distribution of values in mineral deposits. The simpler models are based on the assumption that the values are randomly distributed. Classical statistical methods, based on this assumption of the random distribution of values, are used to analyse mineral deposits to which these models apply or are assumed to apply. In all mineral deposits, however, one recognizes the presence of areas where the values are higher or lower than elsewhere. Also, the values of two samples in a mineral deposit are more likely to be similar if the samples are taken close together than if they are taken far apart. This indicates that there exists a degree of correlation between sample values, and that this correlation is a function of the distance between the samples. Models have been developed which take this correlation into consideration, with the degree of correlation between sample values being usually measured by the semivariogram function. In these models the fact that two samples taken next to each other will most probably not have the same value, must also be considered; even for very short distances the correlations are usually not perfect and a purely random component is present in the value distribution. The mathematical models will therefore assume the presence of two sources of variability in the values: a correlated component and a random component. Finally, one must consider the particular and very common case of mineral deposits in which the values present a systematic variation in space. This variation is usually referred to as a drift, or a trend. For example, the grade of an ore body may increase with depth of the ore, or it may decrease when one moves away from a central volcanic pipe. The earlier models did not give a satisfactory representation of drifts, and more complex models have been developed, in which three sources of variation are represented. These models are made up of: a deterministic component, a correlated component, and a random component. The deterministic component is used as a model of the drift. The correlated component explains regular changes in values which are not represented by the drift. The random component represents variations which cannot be explained by any of the above factors. The simpler models, based on the assumption of a single random component, will be described first (Chapter 2). The models based on the hypothesis of the superimposition of a correlated component on a random component, will then be analysed in detail. These models are most commonly used in the analysis of mineral deposits (Chapters 3-1 I). Finally, how to deal with the presence of a drift will be briefly described (Chapter 12). This book has been written essentially for students in mining engineering and for mining engineers who are interested in the background to the theory of geostatistics as well as its practical applications. The assumption is made that the reader has an elementary knowledge of statistics. Some knowledge of linear algebra is useful in part of Chapter 9, and is necessary to read Chapter 12. A proof is given of all the equations related to geostatistics, and which are not usually found in elementary textbooks on statistics. Understanding of these proofs is not necessary for practical application of the theory, and the reader may wish to skip them on a first reading, concentrating attention on the numerous simple practical examples given. Although the theoretical geostatistician will not find much new material in this publication, it is anticipated that he will develop some interest in the practical approach chosen to prove the various geostatistical equations. Many people and institutions contributed to the preparation and completion of this work. I am much indebted to Dr D. G. Krige and the Anglo Transvaal Consolidated In- vestment Company Limited, who gave me the opportunity to spend a considerable amount of time working on geostatistical problems, both theoretical and practical, during the time that I was in their employment. Dr Krige contributed greatly in developing my interest in studying both the theory and practice of geostatistics, always insisting that a correct balance be kept between theory and practice. I am grateful to Professor H. M. Wells and the Mining Department of the University of the Witwatersrand for inviting me to give lectures in a post- graduate course on geostatistics. The notes 1 prepared for that course became the foundation of the present work. I am also indebted to the Centre de Geostatistique of the École des Mines de Paris, where I received my first formal education in geostatistics during a summer course given by Charles Huijbregts, after many years of my lonely plodding through the published literature. Many graphs in the present volume are reproduced with the permission of the Centre de Geostatistique. The Department of Metallurgical and Mineral Engineering of the University of Wisconsin-Madison has also contributed in making this work possible, by allowing me to spend a considerable amount of time and resources in the writing, typing, and correcting of successive drafts. I am thankful to Lynn D. Kendall, who typed the entire manuscript under constant pressure.

Jan 1, 1978

By T.R. Stacey

Physical testing of rock support for rockbursting conditions has been carried out for over 40 years. A review of this testing shows that it has been mostly component based, rather than actually testing support systems. Further, it is concluded that none of the testing is truly representative of rockburst loading in a similitude sense. Similitude conditions are not achievable, mainly because the real conditions in a rockburst event, such as seismic source location and magnitude, wave frequencies, amplitudes, and interactions, are not all known. Because such information is not available, and because the results of all testing carried out to date have not been able to define, for support design purposes, the capacity of support systems, ongoing physical testing of rockburst support systems is essential. It is essential that the test should simulate, or actually take place in, a supported rock excavation. A rock support system is a combination of individual support components that work together to retain and contain the rock. In doing this, the components are subjected to loading by the rock and to interactive loading between one component and another. It is necessary to prove the capacity of such rockburst support systems by subjecting them to severe loading, as in direct blasting. The direct blasting approach, pioneered more than 40 years ago, probably still provides the greatest validity as a significant test of rockburst support capabilities, even though it does not simulate a rockburst. Direct blast testing of rockburst support systems in a surface environment, such as in a quarry or on exposed rock cutting surfaces, could represent a practical development of the approach, facilitating the execution and monitoring of tests.

Jan 1, 2012

By E. -U. Reuther

To improve the climatic environment underground the German coal mining industry has installed refrigerators with a capacity of 300 MW cooling power. One third of this total is accounted for by central refrigeration plants with more than 5 MW capacity. The central refrigeration comprises three stages: mechanical cooling, transport and distribution of chilled water via pipelines and exchange of heat in the air coolers. Of these three stages only the mechanical generation of chilled water has been resolved satisfactorily. Transport (including distribution) and exchange require further technical research. The Mining Institute of the Aachen University of Technology has done extensive studies on this subject, concentrating on controlling the flow of chilled water according to demand in the mine. Additionally, the favourable exchange in the air coolers is taken into account. The studies are based on a model mine which simulates the conditions in deep coal mines in Germany. The simulation program represents firstly the condition of the mine air at the location of the air being cooled, then the heat exchange in the air cooler itself, and finally the transport of the chilled water from the central refrigerator to the consumer. The program also reproduces the seasonal and operational changes in conditions of mine air as well as different air cooler designs. A regulated distribution of chilled water is also suggested. Finally, estimations of costs can be made as an aid for further planning.

Jan 1, 1987

By A. G. Royle

The reasons for constructing a model of an orebody for which both the parent population and the sampling distribution are known are discussed and details of the model are given. A brief summary of Matheron's theory of regionalized variables is included, with particular reference to his kriging estimators and their variances. Using the model, a comparison is then made between these kriging estimators and other commonly-used estimators. These are compared to find the one with the minimum variance, and an analysis is included to investigate whether any of the estimators is biased. Also included is a comparison between theoretical and experimental variances for some of the estimators.

Jan 1, 2014

By M. Khodadazadeh, K. G. van den Boogaart, T. Leißner, L. Tusa, J. Gutzmer, R. Gloaguen, M. Kern

Ore sorting is a technology that is increasingly used to process primary raw materials. Timeconsuming and expensive empirical state-of-the-art test work is carried out to assess whether the use of ore sorting for the enrichment of a particular ore makes technical and economic sense. With the innovative simulation-based approach presented here, it is possible to direct the selection of a suitable sensor based on quantitative mineralogical and textural data, thus avoiding much of the empirical studies. Required data can be collected quickly and cost-effectively using available methods of automated mineralogy. The obtained parameters such as mineral grain size distribution, modal mineralogy, mineral area and mineral density distribution have been utilised in this study to simulate the prospects of success of ore sorting, applying different types of sensors. Empirical tests with commercially available sensor systems have been conducted to experimentally validate the predictions of the simulations. The estimation of the target mineral grade can be further optimised by the use of machine learning algorithms for the integration of automated mineralogy data and sensor data. The approach can easily be adapted to other types of raw materials and thus has great potential to become a key technology for the optimisation of processing experiments. Keywords: Automated mineralogy, hyperspectral imaging, machine learning, ore sorting

Jan 1, 2020

By S. J. Bluhm

Recommendations are given for the design of multi-stage spray chambers in which air flows horizontally through water sprays. A detailed description is given of one such spray chamber at Hartebeestfontein gold mine that is used for bulk cooling of the incoming ventilation air, and that has been constructed according to these recommendations. A method for predicting the performance parameters of horizontal multi-stage spray chambers is given, and the method is illustrated with some examples. The method is equally applicable to spray chambers in which the water, rather than the air, is cooled.

Jan 1, 1979

By C. Tokoro, T. Kato

The objective of this study is to improve the shrinking core model for particles with a particle size distribution and clarify the activation mechanism of rare earth elements dissolution by mechanochemical activation. To achieve this objective, the method, called the doubly integrated micromodel, was applied for a shrinking core model in this study. This method consists of integrating the batch leaching rate of a single particle over the distribution of both feed particle size and reactor residence time to arrive at the fraction unleached. The usefulness of the shrinking core model improved by doubly integrated micromodel was confirmed by comparing it with a shrinking core model. To apply the shrinking core model improved by doubly integrated micromodel to rare earth elements dissolution from weathered residual rare earth ores by mechanochemical activation, leaching experiments were performed for grinding samples by planetary ball milling. The reaction equation of mechanochemical reaction by planetary ball milling for weathered residual rare earth ores was investigated by x-ray absorption fine structure analysis. As a result of a kinetics study, the rate-determining process of leaching changed from reaction to diffusion reaction control by planetary ball milling. Based on these results, it was revealed that the main mechanism of activated rare earth elements dissolution from weathered residual rare earth ores was a mechanochemical reaction by planetary ball milling. Keywords: Mechanochemical reaction, particle size distribution, shrinking core model

Jan 1, 2020

By Z-P. Song, Z-B. Jiang, A-N. Jiang

"In deep excavations under conditions of high ground stress, even hard rock undergoes creep and other time-dependent effects that may result in instability. This study involves an analysis of the typical damage to the surrounding rock in the deep tunnels of the Jinchuan nickel mine in China. In order to understand the creep mechanism, triaxial creep laboratory tests were conducted under high stress. The creep characteristic curves and the Nishihara model parameters of typical rock were obtained from these tests. The numerical simulations, with which the creep model is developed, were performed using Flac3D. The creep model parameters were back-analysed on the basis of field monitoring data. Finally, appropriate support schemes for deep tunnels are suggested. The creep deformation of rock increases with time and axial load. At the same load level, the creep deformation of unloading is greater than that of loading. The simulated deformations of the surrounding rock are coincident with those from in situ monitoring, indicating the applicability of the presented rock creep model in situations of high ground stress. The results constitute a reference for support design in deep mining tunnels.IntroductionThe excavation of underground caverns can alter the natural stress field of the surrounding rock. In deep engineering projects under conditions of high ground stress, even hard rock shows obvious creep properties and time effects (Yao, 2005; Liu, 1994). The time effect of excavation not only causes decompression of the floor and increases sidewall stress, but also leads to stress concentration at the corners of the excavation, which influences the stability of the surrounding rock in a very complex manner.Many researchers have studied the creep properties of rock. ‘Griggs (1939) showed that creep occurs in sandstone, shale, and silty sandstone rock under the load of 12.5%–80% of the failure load.’ Subsequently, much research has been published on rock creep tests, including the uniaxial compressive creep test, the triaxial compressive creep test, the discontinuity rock shearing test, and field tests (Wawersik, 1974; Amadei et al., 1980; Okubo et al., 1991; Jin et al.,1995). A rheological model is used to describe the complex rheological deformation of material. This model is based on various fundamental parameters including rock elasticity, plasticity, and viscous properties, which can be combined into a constitutive model to reflect the rheological properties of various types of rocks. Wang et al. (2014) studied the creep– damage–rupture characteristics of rock salt by applying a creep-damage model. Munson (1997) developed a multi-mechanism constitutive model of creep of polycrystalline rock salt based on steady-state creep and modified it to incorporate transient creep through work hardening and recovery. An elasto-viscoplastic approach was developed to simulate the timedependent nature of the fracture zone (Malan, 1999). Rheological models were proposed to understand the collapse of iron ore mines in Lorraine by Dragan et al. (2003)."

Jan 1, 2016

By C. Musingwini, T. C. Sebutsoe

"The primary objective of any mining business unit is to make profit by extracting, processing, and selling minerals from a particular mineral deposit. It is important to optimize extraction of the mineral resource given time, capital, and space constraints. Mineral extraction is often associated with uncertainty due to variable technical and human factors. Technical factors such as grade distribution, ground conditions, and equipment reliability influence the performance of a mining production system (MPS). The performance of the MPS is also impacted by human factors such as employee skills, health, and attendance. Uncertainty associated with technical and human factors often leads to actual output differing from what was planned. Therefore an in-depth analysis of the significant causes of deviations from the planned outcomes is relevant.In this paper we investigate the empirical relationships between inputs and outputs in a MPS in order to assist management in directing efforts at key production drivers. Once a MPS has been empirically characterized, more effort and resources can be focused on the key decision-making variables (DMVs) in order to meet the planned outcomes.For a typical underground, tabular, narrow-reef conventional drill-andblast platinum mine the face advance, face length mined, number of teams, team efficiencies, and team size have a statistically significant relationship with the centares (m²) produced, which is a key performance indicator (KPI). The production function derived in this study can be used to align the physical, technical, and human factors together to predict the optimal output level. The most significant production lever of the MPS is the face advance. IntroductionA mining production system (MPS) is a result of an iterative process of design, planning, and optimization of mining input variables and decision-making variables (DMVs). The MPS exists within the mineral extraction link of the complete mining value chain. It represents the stage where mining companies have the opportunity, through production, to start generating returns on the investments undertaken. Returns on shareholders’ investments can be realized at this stage."

Jan 1, 2017

By B. Mienie, J. F. Morrison, C. H. Wyndham, M. J. Martinson, N. B. Strydom

Contribution to discussion M. J. Martinson (Associate Member)*: Application of the simple selection test mentioned in the penultimate paragraph of the paper (reference to a more detailed description of the test would not have been inappropriate) should assist South African mining engineers to make more efficient and possibly more humane use of available labour. The authors mention that the test 'is in use in the gold mining industry', and it would be interesting to know the extent of this usage in terms either of number of mines or as a percentage of the total number of labourers employed on 'heavy' tasks. The paper would have been of greater interest had the authors attempted to relate their findings to the statistics of labour usage in gold mines, members of the Chamber, and to the 'heavy' tasks in particular. On this the authors confine themselves in their Introduction to the somewhat ambiguous statement that 'A relatively high proportion of the 360,000 Bantu mine labourers are employed underground on manual tasks'. To appreciate the practical significance of the authors' work one needs a detailed breakdown of the underground labour force by occupations, together with estimates of quantities such as: (a) total stope and development tonnages moved manually per shift: by shovelling by shovel/tram by tramming together with distances and any other relevant data; (b) theoretical energy expenditures required to accomplish (a); (c) actual energy expenditures required to a accomplish (a); (d) energy theoretically available to accomplish (a) based on 50 per cent of the total maximum oxygen intakes of the labour force allocated to these tasks. Analyses of this sort on an industry basis might demarcate the more promising areas for detailed investigation. Stress is often laid on the need for increased mechanisation to increase stope productivity, but it may be questioned if enough is known about the characteristics, performance and potential of our underground labour force to warrant blind acceptance of stope mechanization as the industry's panacea. Provided that health and safety are not compromised, the ultimate yardstick for assessing any form of stoping is profitability; obviously the industry must investigate very thoroughly any form of mechanization which might reasonably be expected to increase profitability, but it should also be remembered that a radical overhaul of existing systems-which have after all evolved by trial and error over the past 75

Jan 11, 1968

In a gold CIP/CIL circuit, adding new circuit water to a final tails thickener and then directing the thickener overflow to the head of the circuit significantly reduces the cyanide concentration at the TSF discharge spigot. The new water is used to dilute the final tails. The dilution and recycle of the tails in this manner may have the added benefit of reducing reagent consumption, in some cases to the point where the reagent savings justify the expenditure and Cyanide Code compliance is achieved at no extra cost. Potential circuit configurations and water recycles can be modelled by splicing a flowsheeting and thermodynamics package, such that changes to the composition of the water can be reduced to chemical speciation with a Gibbs free energy minimisation enabling a complete understanding of the circuit water chemistry.

Jan 1, 2007

By R. G. B. Pickering, H. H. Haase

Few innovations in the South African gold mining industry could have a more significant impact on improving profitability than non-explosive methods of mining. The far-reaching benefits of this approach are explained and the progress made through development work funded by the industry is described. Three methods are considered: impact ripping, sloting with abrasive waterjets and diamond saws, and breaking from drilled holes with waterguns and electric pulses. Of these methods, impact ripping is the most advanced and requires only minor refinements before production applications can be initiated. Although this method, based on current performance, is expected to yield considerably increased profits, it is predicted that it could perform several times better than at present with further development. The other methods of rockbreaking also show potential. Waterguns and electric rockbreaking would be attractive in unfractured rock for difficult geological conditions, and for confined areas such as gullies and development ends. Methods based on sloting could be combined with any of the other methods to enhance their effectiveness and thus yield further improvements in profitability; also, they could be used to advantage in specific applications. Al the discussed methods, therefore, deserve to be developed further.

Jan 1, 1991

By J. M. Eriksen, T. Rosenqvist, H. Elstad, S. Seim, A. Hildal

The titanium industry demands high titanium feedstock for their processes. As rutile resources are limited, ilmenite has become the most important titanium source. Ilmenite smelting is an important part of upgrading ilmenite to high titanium feedstock. In the year 2005, upgrading of ilmenite by smelting, producing a slag with 80-90% Ti02, accounted for approximately 40% of the feedstock to the Ti02 pigment industry. In ilmenite smelting molten iron is produced together with a slag, which in addition to titanium oxide contains iron oxide as well as various impurity oxides of manganese, magnesium, silicon, aluminium, etc. Titanium is present mainly as Ti02 but partly also as Ti203. It is known that the concentration of Ti203 increases with decreasing concentration of FeO, but this dependency is affected by the nature of the impurity oxides. Sigurdson and Moore1 showed that slags high in CaO and MgO had a lower concentration of Ti203 than slags high in Si02 for the same amount of FeO. Pistorius2 showed in 2002 that if the impurity oxides MnO and MgO are counted, on a molar basis, as FeO, and Cr203, V203 and part of the A1203*), as Ti203, industrial slags will have a composition close to that of the M305 phase, where M denotes the elements Ti, Fe, Mn, etc, and which has the pseudobrookite structure. However, for equilibrium with liquid metallic iron he computed, by means of a computer program FACT, that the slag equilibrium composition is represented by a curved line in the ternary phase diagram FeTi03-TiO2-Ti2O3, but with a significantly higher concentration of trivalent titanium than actually observed in the industrial slags, and he discussed various mechanisms by which the slag during solidification could change its composition to a lower Ti203 content, but the possibility that the computed compositions might be wrong was not mentioned. Trane113 found in her research on titaniferous silicate slags a lower Ti3+/ Ti4+ ratio than that which is computed by the FACT program. The purpose of the present investigation was to study the equilibrium composition of Fe-Ti-O slag in equilibrium with liquid metallic iron, as well as the effects of impurities such as Si02, CaO and MgO on the slag composition. Industrial slags from TINFOS Titan and Iron KS (TTI) will also be discussed and their compositions will be plotted in the ternary phase diagram. These are slags produced from Tellnes ilmenite, quite high in magnesium and iron. Experimental In the past, equilibrium studies of the Fe-Ti-O slag system were difficult when it came to producing the slag. It was difficult to find a suitable crucible for the meltings. The highly corrosive slag would attack all known refractories, and the metal would attack metallic crucibles such as those made of platinum and molybdenum. In the present investigations the slag-metal combination was prepared in a high-frequency induction furnace with a vertically segmented water-cooled copper crucible. The crucible is shown in Figures 1 a and 1b. It was designed by the Australian Nuclear Science and Technology Organisation (ANSTO) with power from a 75 kVA high-frequency generator (750 kHz) supplied by Farfield Electronics Pty, Ltd, Australia. This enabled the slag-metal combination to be melted with turbulent stirring and is expected to be isothermal. The entire temperature gradient

Jan 1, 2007