Search Documents

By Neeraj Saxena, Madhu Huggahalli, Jerry Bednarski, Ken Grieshaber, David Stoffel

"The use of oxy-fuel burners in secondary aluminum melting applications offers several advantages over air-fuel burners including reduced fuel consumption, faster charge to tap times and lower NOx emissions. Their successful, safe and economical use in a furnace depends on several factors and considerations such as burner and flue placement, metal circulation, charge practices and the type of refractory used in the furnace.These factors required in the successful conversion of furnaces from air-fuel to oxy-fuel based burners are discussed in this paper. Critical parameters are identified and examined using computational fluid dynamics (CFD) simulations. Parameters are estimated via laboratory testing and validated through trials performed at commercial installations. Guidelines and a simplified approach to estimate a priori the economic impact of converting from air-fuel to oxy-fuel are presentedThe final products of this research are simplified and validated tools for modeling aluminum furnaces. These include improved heat and energy balance models and parameterized CFD solutions which provide rapid customization to individual furnace configurations. These tools provide field engineers with immediate and accurate predictions of performance, allowing for repeated precise scenario analyses."

Jan 1, 2001

By K. Kalgraf

The motion and stability of the metal/electrolyte interface is determined by Navier-Stokes equation. The equation may be decomposed into two independent equations by taking the divergence and curl of Navier-Stokes equation. The divergence determines the motion and stability of magneto-gravitational waves. The curl of Navier-Stokes equation gives an equation containing only velocity and magnetic field, and states that the magnetic field is driving a circulation pattern in the cell, and this magneto-flow has its own stability conditions. The waves are called Alfven waves, and their stability properties are related only to magnetic field and geometry.

Jan 1, 2001

By W. G. Davenport

Chapter 4 examined Inco (oxygen) flash smelting of pure chalcopyrite concentrate. It showed that the oxygen and flux requirements for auto thermal smelting are readily determined by a matrix calculation. This chapter extends this matrix approach by applying it to a multi-mineral concentrate.

Jan 1, 2001

By A. P. Pavia

Investigation on profitable ways to recover significant quantities of silver remaining dissolved in highly concentrated chloride solutions frequently pass through the development of solvent extraction (SX) routes. In this work, the general pathways involving Ag(I) extraction by dibutylthiourea (DBT) or tetraethylthiuram disulfide (disulfiram, DSF), from concentrated acidic chloride solutions, are investigated. With this purpose, the dependence of Ag(I) distribution coefficient on ligand and chloride concentrations has been analysed. Furthermore, in order to check whether there is some protonation at any site of DBT or DSF after contact with adequate chloride phases, lH NMR has also been used. The collected spectroscopic data suggest that anionic Ag(I) extraction reactions are not likely to occur. Based on the overall results achieved, proposals for the most probable Ag(I) extraction reactions are made.

Jan 1, 2001

By C. Takano

The integrated iron and steelmaking industry generates large amounts and many types of solid dusts and sludge, at different phases of the production. Within these are: sludge of the Blast Furnace; coke fines; fine and coarse fractions from oxygen converters; sludge from water treatment at rolling mill unit; and others. In this paper the above dusts and sludge were physically and chemically characterized. The obtained results allowed defining self-reducing pellets using these materials. The high temperature behavior of the pellets were evaluated. Good results, with no decrepitation and swelling, and high yield of reduction, show that it is feasible to recycle them, as self-reducing pellets, in oxygen steelmaking converter.

Jan 1, 2001

By W. G. Davenport

A flash furnace can be run many different ways. It can, for example, be operated with: (a) highly 02"enriched, ambient temperature blast and no fossil fuel (b) considerable fossil fuel and hot, slightly Or enriched blast. It can also produce different grades of matte. With all these possibilities, the flash furnace operator can benefit from any technique that will indicate the 'best' way to operate his furnace. One such technique is Excel's 'Solver'optimizationroutine#. This chapter shows how our matrices and Excers optimization.abllity can be used to optimize a flash furnace operation. The chapter sets three optimization goals: (a) minimum industrial oxygen + oil cost, $ per tonne of concentrate (b) maximum smelting rate, tonnes of concentrate per hour (c) maximum profit, $ per hour.

Jan 1, 2001

By Marco V. Ginatta

Current world production, 60'000 ton/y, is exceedingly too small for titanium's extraordinary combination of favorable properties; it should be 1,000,000 ton/y (7% of stainless-steel). Prices that competitively sustain that sales volume are achievable only with electrolytic production, as it is for all other commercial nonferrous metals. But titanium does not have its commercial electrolytic plants yet, because of producers decisions and strategies, scientists works, industrial problems with chloride process, lack of consumers sponsors. Fluoride high temperature process has the advantages of aluminum electrolysis, plus other favorable characteristics specific to titanium and its feed material. One electrolytic titanium potroom replaces several different plants used for sponge production. Production of titanium ingots with zero defect is achieved. The solidified cathode rectangular slabs are suitable for direct rolling.

Jan 1, 2001

By K A. Pericieous

"Magnetic fields have many actual applications in the metals processing industry. Externally applied magnetic fields give rise to electromagnetic (Lorentz) forces formed by the cross product JxB, between the induced current density J and the magnetic field density B. When the metal is in liquid form, the Lorentz force generates motion in the fluid which in applications of practical interest becomes turbulent. In modelling terms, the Lorentz force appears as a source in the momentum equations. In addition, the induced current generates heat (Joule heating) in the metal that is in proportion to J2, with a corresponding source of heat in the energy equation. Whether as heat or as a force, these effects represent action at a distance - a most useful attribute when dealing with hot metal. The Lorentz force is used to stir solidifying alloys, pump liquid metal in conduits, dampen the flow in the meniscus of a continuous caster, levitate metal drops, induce artificial gravity conditions in suspensions or contain liquid metal. Elsewhere, the Lorentz force may be a by-product of some other operation, so leading to wave excitation in aluminium electrolysis cells, or altering the shape of the weld pool in arc welding. Joule heating is most commonly used with applied AC fields, to melt metal in induction furnaces.The author and his colleagues have been involved in the modelling of most of these processes in the past decade. Modelling is not however straightforward, since most of the examples mentioned represent genuine multi-physics challenges. There is a strong coupling between the flow field and electromagnetic field. The addition of a dynamically varying metal free surface and the moving solidus front means the flow, heat and electromagnetic fields need to be computed simultaneously. In situations involving metal containment, the metal free surface position is governed by the interplay of gravity, Lorentz force, surface tension and fluid inertia. Since all the interesting effects often happen in thin boundary layers at the surface due to the skin effect, mesh generation and mesh control during the computation become non trivial problems that need to be addressed. This paper presents a review of numerical methods used to model droplet levitation, semi-levitation melting and cold crucible induction melting of metals. The first method is based on spectral collocation techniques and the second is the traditional FV approach. Steps taken to validate the computations and typical transient results are also given."

Jan 1, 2001

By W. G. Davenport

Chapter 1 showed that flash smelting is a key step in extracting copper from sulfide ores. It also showed that there , are two versions of flash smelting: (a) Outokumpu flash smelting, which uses oxygen-enriched air blast and injects dry feed downwards into a hot furnace (b) lncoflash? smelting, which uses industrial oxygen blast and injects dry feed horizontally into a hot furnace. This chapter describes Outokumpu flash smelting.

Jan 1, 2001

By Vaughan R. Voller

"The wide range of length and time scales found in solidification processes are outlined and discussed. Methods for Direct Microstructure Simulation (DMS) are introduced. Key features in sharp interface and phase field models are presented. Concepts in micro-macro solidification modeling are covered. Basic details of microstructure and segregation models are provided. A description of a recent segregation micro-macro model is presented in detail.IntroductionA solidification process involves the containment of a liquid region, which, due to cooling, transforms over time to form a fully solid product. The development of the solid microstructure and the segregation of solute components during this transformation are controlled by phenomena that occur across a wide range of length and time scales. A ""successful"" model of a given solidification process needs to account for these phenomena. Standard models of solidification employ a discrete grid of node points and associated volume elements in the special domain, with a discrete time step to track the transient changes of nodal values, describing the composition and microstructure. Through the use of such models significant progress has been made towards a complete understanding of solidification process and phenomena (see the conference proceedings in Refs [1-9]). As solidification models become more sophisticated, however, they all run into the problem of resolving the full range of disparate space and time scales.Two approaches can deal with this problem1. Direct Microstructure Simulation (DMS) where the grid size and time step are chosen to scale with the smallest length and time scales of the given problem.2. Micro-macro modeling where the grid size and time step scale with the macroscopic process scale and phenomena at smaller scales are accounted for via sub-grid modeling and volume averaging.Currently full DMS of solidification is beyond the reach of available computer power and the application and development of micro-macro models is an important and critical research area."

Jan 1, 2001

By W. G. Davenport

The Cu content of industrial flash furnace concentrates varies considerably. It is usually around 30% Cu (Tables 2.1, 3.1) but it may be as high as 55% Cu and as low as 20% Cu. The mineralogy of flash furnace concentrates also varies. The most common Cu mineral is chalcopyrite (CuFeSz), but many concentrates contain bornite (CuSFeS4), chalcocite (CU2S), covellite (CuS) and even native copper (CUO). Pyrite (FeSz) :is also present in many concentrates. This chapter examines how concentrate composition affects flash smelting -with specific reference to 'low-Cu-grade' FeS2-CuFeSz concentrates. Chapter 14 does the same with 'high-Cu-grade' CuzS-CuFeSz?concentrates.

Jan 1, 2001

By W. G. Davenport

This chapter demonstrates how our matrices can be used for controlling a flash furnace. Specifically, it shows how we can determine the adjustments in industrial oxygen, air, oil and flux input rates that will; (a) correctly alter product temperatures and compositions to newly targeted values (b), maintain set-point product temperatures and compositions with varying feed compositions and feed rates. It also discusses corrective actions that might be taken when furnace temperature and matte grade unexpectedly change.

Jan 1, 2001

By W. G. Davenport

Flash smelting is used. mainly for making molten Cu-rich matte* from fine Cu-Fe-S** flotation concentrates. It blows dried concentrate particles into a hot fumace, surrounds them with 01-rich gas and allows them to react. The results are: (a) partial oxidation of Fe and S from the concentrate (b) generation of heat (c) melting of the oxidized particles to form molten Cu-rich matte, ~65% Cu and Cu-barren slag, ~1 % Cu. The melted particles fall to the hearth of the flash furnace where they fonn matte and slag layers, The matte is tapped from the furnace and oxidation-converted to metallic copper. The slag is skimmed, treated for Cu recovery then stockpiled or sold. Figures 1.1 and 1.2 show industrial flash furnaces and the position of flash smelting in the coppermaking flowsheet. ,Fig. 1.3 shows the locations of flash furnaces around the world.

Jan 1, 2001

By W. G. Davenport

Most pre-t 988 Outokumpu flash furnaces were built with blast heaters. Many continue to use them. Blast temperatures in these smelters· are typically 400 to 700 K (Table 2. 1) This chapter: (a) shows how hot blast is included in our calculations (b) evaluates the effects orhot blast onauthothermal65% Cu matte production.

Jan 1, 2001

By W. G. Davenport

All Outokumpu flash furnaces burn some hydrocarbon fuel. Hydrocarbon combustion supplements heat from Fe and S oxidation and provides local heating in cool; parts of the furnace. Adjustment of hydrocarbon fuel combustion rate is an effective way of controlling furnace temperature. More oxygen and more oxidation of Fe?and S in the flash furnace continue to diminish the need for1hydrocarbon fuel. Still, as Tables 2.1 and 3.1 show, all flash furnaces use it. Oil is the most common fuel -natural gas, coal and coke are also used. This chapter demonstrates how hydrocarbon fuel is included in our flash furnace matrix. It shows that: (a) three new variables, mass hydrocarbonfoel. mass CO2 in offgas and mass H2O in offgas are inserted (b) carbon and hydrogen balances are added (c) hydrocarbon fuel quantity is specified (d) ,the oxygen balance is modified to include CO2 and H2O (e) the enthalpy balance is modified to include hydrocarbon fuel, CO2 and H20.

Jan 1, 2001

By Christian Redl

"The Lattice Boltzmann Method (LBM) solves fluid dynamics problems based on a discrete mesoscopic approach. It is based on the Lattice Boltzmann equation which is a special discretisation of the continuous Boltzmann equation. Its principle advantage is that it can handle complex physical phenomena (like interaction between different components and surface effects), complicated boundary and initial conditions very well, free from many gridding and stability constraints that plague conventional numerical methods used for fluid flow simulations. Complex structures like porous media can be resolved directly. The LBM algorithm has a simple structure and acts locally, which is favourable for parallel computing. Its potential for metallurgical process simulation is demonstrated in the present study which is concerned with the simulation of the fluid flow in a copper winning electrolysis cell. The electrolyte is modelled as the carrier fluid and the oxygen is considered via a so called active scalar equation. The modelling of buoancy, momentum exchange and the free surface requires special effort. The results of the simulation are validated against experimental data obtained by Laser-Doppler-Anemometry. Despite the simplifications made, good agreement between the simulation and the experiments was found.IntroductionThe hydrodynamics in an electrometallurgical cell is a complex, highly non-linear phenomenon. It includes the simulation of a two component system with high density ratio, gravity driven flow and a free surface. A general solution to this problem is currently unknown.The flow field is a result of the combination of various hydrodynamic effects [18]. Metal deposition at the cathode yields in a decrease in the ion concentration, which reduces the density of the electrolyte. At the anode molecular oxygen is formed as a consequence of the oxidation of water molecules. The oxygen is solved in the electrolyte and again the electrolyte density is decreased. Both effects lead to a free upwards directed convective flow, refered to as natural convection."

Jan 1, 2001

By W. G. Davenport

This chapter shows how flash furnace industrial oxygen and oil requirements are affected by different: (a) product temperatures (b) conductive, convective and radiative heat losses (c) electrical energy inputs (d) hydrocarbon fuels. The calculations of the chapter are all for production of 65% Cu matte from pure CuFeS2 concentrate. Blast temperature is? 298 K throughout.

Jan 1, 2001

By W. G. Davenport

Chapter 11 determined the combinations of industrial oxygen, oil and blast preheat that will steadily produce 65% Cu matte from pure chalcopyrite concentrate. It showed that the . combinations are described by a flat plane. It also showed that the minimum energy requirement for making 65% Cu matte occurs with maximum oxygen enrichment, zero oil and zero blast preheat. This chapter examines the, effect of matte grade on industrial oxygen, oil and preheat requirements.

Jan 1, 2001

By James W. Reeves

DuPont was the titanium metal pioneer who put the greatest effort into both R&D and commercial sponge and powder metallurgy production. This was concentrated in the period 1945 through 1963. The commercial results were development of the chloride process for TiCl4 using ilmenite, development and improvement of the Kroll process, development of a modified Hunter process to make titanium powder and commercial development of a direct powder process. The powder metallurgical process failed because of the inability to meet powder chloride specifications. All this work remained a trade secret until published in NMAB -392 by Mahla in 1983. There were other processes explored by DuPont, both improved sponge and powder processes. Much of this work has been continued by others but none, as of yet, resulting in new improved titanium metal capacity.

Jan 1, 2001

By Noritaka Horii, Katsuyoshi Kakinuma, Hiroshi Yamamura

"ZrO2 and Y2O 3-doped ZrO2 gel film were obtained by spin coating of sol from Zr-n-propoxide, n-propanol, acetylacetone, nitric acid, and water on a silica glass as a substrate. ·Dried gel films crystallized to monoclinic, tetragonal, cubic and their mixed phases, depending on the chemical composition of sol, .firing temperature, and coating cycles. ZrO2 films with highly preferential orientation were successfully obtained from a solution containing a large amount of n-propanol. The highly oriented ZrO2 film was found to have high transparency compared with non-oriented 'films. Electrical conductivity of 8 mol% Y2O 3 doped ZrO2 film was also measured.Introduction3 mol% Y2O 3 doped ZrO2, which has a tetragonal symmetry is ·well known to have a strong mechanical property and a high toughness. (1) On the other hand, 8 mol % Y 2O 3-doped ZrO2 is high oxide-ion conducting material and is applied to oxygen sensor, oxygen pump, and solid oxide fuel cell (SOFC). (2) Sol-gel methods using metallic alkoxide are widely applied for synthesizing ceramics such as high-purity fine particles, thin film, and fiber. (3) For ceramics films, many reports have been made on SiO2, (4) TiO2, (5) and PZT. (6) Several case studies have been done on zirconia film based on the sol-gel method, (7) - (11) but very few on preferentially oriented film and on systematic Y zO3 doping. Almost zirconia ceramic· films reported were prepared by means of dip-coating. Kamiya et al. (12) and Ganguli et al (13) converted zirconia into fiber. Sol from Zr alkoxide is usually very unstable, and a stabilizing agent such as glycol or ß-keton is used beforehand (7) (14)."

Jan 1, 2000