Light Metals 2016

Low-grade bauxites and non-bauxite resources are widely distributed, which are suitable to extract alumina by the lime sinter process. The application of the lime sinter process is restricted by the high consumption of calcium, the large amount of slag and the poor alumina leaching property of clinkers. An improved lime sinter process based on the dry-sintering is proposed by adding Na₂O-containing compounds to decrease CaO addition and increase alumina leaching property. The molar ratios of CaO to Al₂O₃ and Na₂O to Al₂O₃ are 0.6 and 0.4, respectively. Na₄Ca₃(AlO₂)₁₀ as well as trace amounts of CaO•Al₂O₃ and 12CaO•7Al₂O₃ are formed after sintering at 1200°C for 30min in CaO-Al₂O₃-Na₂O system, the alumina leaching efficiency of which is above 90%. The minerals after sintering in CaO-Al₂O₃-SiO₂-Na₂O system include Na₄Ca₃(AlO₂)₁₀, 12CaO•7Al₂O₃, 2CaO•SiO₂ and 2CaO•Al₂O₃•SiO₂. The leaching efficiency of clinkers increases as the sintering temperature increases to 1300°C.

High content of silica results in high soda consumption, which limits the use of high-silica bauxite in the production of Bayer alumina. In this work,the main flotation parameters of high silica bauxite (A/S 5.0) including particle size, pulp pH, collector dosage and dispersant dosage were studied. With an optimal condition of 33% solid density, particle size of -74 μm above 93%, sodium oleate dosage 2200 g•t−1, Na₂CO₃ dosage 2000 g•t−1, sodium hexametaphosphate dosage 30 g•t−1, slurry pH 9.4 and agitation speed 2280 rpm, the recovery of aluminum minerals of 56%, the A/S of concentrate of 12.2, and the separation index of 0.4 can be obtained.

With the rapid development of China alumina production, efficient utilization of plentiful low grade high-iron bauxite has received widespread attention in the aluminum industry. Based on this, “calcification-carbonization-direct reduction” method is proposed for extracting effectively valuable metals Al and Fe from high-iron bauxite. The research results indicate that: after the calcification-carbonization process for high-iron bauxite, leaching rate of Al₂O₃ can reach 83.67%, higher than extraction rate of Al₂O₃ through Bayer method. Then, direct reduction of calcification-carbonization residue is investigated. Metallization ratio of Fe is 80.5% when reduction temperature and time is 1050°C and 2h respectively. Moreover, metallization ratio of Fe can reach 93.6% with the addition of additive A. After the “calcification-carbonization-direct reduction” process, valuable metals Al and Fe can be extracted effectively from high-iron bauxite. The main component of final residue is CaCO₃ and CaSiO₃, which can be directly applicable to the cement industry.

The main problem is facing the wide use sintering process of the aluminum refining raw material (nepheline, ash from coal, etc.) are a high power consumption and an expensiveness of alumina. Improvements are aimed at reducing the consumption of heat and more efficient use of secondary energy resources. Reducing the cost of alumina is expected to be reached by inclusion of technogenic raw materials, enhancing the capacity of production equipment and reducing the use of expensive chemical products. It is shown that the production of high demand products and the creation of new processes for their preparation for alumina refineries can not only increase the profitability of processing of raw materials, but also eliminate or reduce process steps, which produces the main amount of waste.

In this paper we investigate the process of autoclave extraction of boehmite-kaolinite bauxite for Severoonezhsk deposits by hydrochloric acid. The effect of temperature and duration of the process on transition of aluminum into solution has been studied. The apparent activation energy was defined for the process. The influence of hydrochloric acid concentration on alumina transition in solution was shown. Determined the apparent degree of the process on reagents. The obtained results allow to allocate 3 limiting stages of the process: 150 – 160°C — a chemical reaction (E _a = 70.88 kJ/mol), 160 – 170°C — intermediate region (E _a = 40.22 kJ/mol), 170 – 180°C — diffusion area (E _a = 22.66 kJ/mol). Thermodynamic modeling of physicochemical processes underlying the preparation of aluminum containing solutions has been performed. It has been demonstrated that the major aluminum species in solution are aluminum chlorides and hydroxychlorides.

The adsorption of sodium formate on goethite or hematite surface was investigated by FT-IR and XPS, respectively. Periodic plane-wave density functional theory (DFT) calculations were performed on models of carboxylate surface complexes on the goethite (101), (100) surfaces or hematite (001) surface. The photoemission core-level shifts (CLS) of the adsorbed surface iron sites calculated with periodic interfacial structures were compared with the experimental XPS. FT-IR results reveal that the interfacial structures may be assigned to bidentate mononuclear (BM) complexes formed via bonding of one surface iron atom of goethite or hematite to both oxygen atoms of carboxylate group. In agreement with the adsorption energies determined by DFT calculations, the calculated results show that only the calculated Fe2p CLS of the adsorbed surface iron sites on goethite (100) and hematite (001) surfaces are correspondingly in accordance with the experimental observed Fe2p CLS, which confirms that the BM structure occurs upon adsorption.

The bauxite in the northern region of China belongs to diasporic type ore with high silica. The main silica minerals in the bauxite is kaolinite, illite and chamosite. Technological investigations were carried out based on the Bayer process. The presence of high silica causes high bound-soda losses in the red mud and the decrease of leaching rate. The dissolution behavior of different silicate minerals has obvious effect on the digestion of bauxite. Predesilication characteristic of the bauxite, the effects of dissolution temperature and retention time on Al2O3 digestion efficiency were determined. Experiment results showed that of chamosite and illite mineral characteristics and dissolution characteristics of chamosite helped to lower alkali consumption to improve the digestion rate of bauxite.

The hydrolysis loss of digested alumina from bauxite reentering into the bauxite residue has always been the research focus during the Bayer process. The stability of sodium aluminate solution after digestion with different lime additions during the dilution process from a diasporic bauxite was studied, and the mechanism by which the lime addition affects was also discussed. The molar ratio, silica index and hydrolysis degree of diluted liquor after digestion are highly affected by the lime addition, and the corresponding alumina extraction efficiency of bauxite and the Na₂O to SiO₂ ratio of bauxite residue are changed. The stability of diluted liquor is firstly increased by the lime addition below 15% of the bauxite, and then decreased as the lime addition increases. The mineral content and proportion of desilication products precipitating during digestion are the main factors affecting the stability of sodium aluminate solution after digestion.

Aim at improvement leaching reaction and heat transfer efficiency, A novel self-stirring tubular reactor driven by pressure energy used in bauxite digestion process was presented originally by Northeastern University, which combined the advantages of autoclaves and conventional tubular digestion equipment. Quasi-logarithmic equation between the stirring speed and the various factors was established by dimensional analysis method. The results demonstrated that higher pressure promoted the appearance of more obvious turbulence and improved the mixing effect of media in the reactor. The quasi-logarithmic equation between the stirring speed and the various factors was: N = 356.7053ρ−0.4322μ−0.13546P0.56773H−0.86454

A study was performed on the leaching kinetics to recover the alumina from Tricalcium aluminate (TCA) generated as waste after removal of suspended red mud particles in the Bayer process by different concentrations of Na₂CO₃ solution at different temperatures and solid:liquid ratios. The influences of initial sodium carbonate and stirring speed on reaction rate constant were examined and the corresponding apparent activation energy was calculated. The results indicate that the leaching process is a first order reaction, and the reaction rate constant 8.64×10–6 mol/(L.m). Optimization of recovery and minimization of Na₂CO₃ in liquor were studied, and the dissolution rate of TCA and conversion of CaCO₃ also increases. The apparent activation energy is 52.86 kJ/mol, and temperature has great influence on the leaching rate of TCA. The leaching rate is controlled by diffusion of reacting reagents in the solid product layer. A higher leaching rate of TCA is obtained by higher temperature and/or lower solid g/l during leaching process..

Low-temperature desilication process was investigated during alkali aluminosilicate raw materials treatment by mode of production of alumina by sintering method at industrial conditions. Nepheline concentrate used as seed for desilication. During the research process, temperature changed from 85 to 100 °C and seeding from 9.6 to 31.0 kg/m3. Accomplishment of desilication at the first stage with use of nepheline concentrate as seed reduces steam consumption for the desilication process and allows to simplify the equipment decision in comparison with autoclave conditions. However, when the non autoclave desilication is applied, the amount of solid phase in a state of circulating slurry substantially increase and contains up to 30 % of the liquid phase in the form of desilicated liquor. The best results achieved at temperatures close to the maximum and amount of seed more than 20 kg/m3. It was established that during low-temperature desilication aluminate liquors dilute with nepheline pulp liquid phase while entering desilication in contrast to autoclave one. This dilution results in aluminate liquor hydrolysis, and the material flows gain which needs increasing volume of settling tanks and equipment; the latter leads to decrease of feeding sinter leaching liquor concentration. As the result of hydrolysis the solid phase after the desilication has in its composition an average amount of mass Al₂O₃ equal to 3,0 % in the type of Al(OH)₃. As well, liquor dilution results in increase of heat consumption for the liquor evaporation in the soda production area.

The synthetic polymer flocculant has been applied to Bayer process for thirty years. In recent years, it is pointed out that the flocculation is related to the presence of other ions. The experimental results show that the effect of mud sedimentation and mud compaction will affect the production capacity and purity of alumina product. The potential of using water absorbing polymer or synthetic dispersant to change the rheological property of the slurry is also summarized in the end. The combination of these effects, with the new bio polymer to improve the primary pour into the washing bath of the overflow liquid clarity and recovery of sodium hydroxide also made a narrative.

The Shell “Prelude” project is embarking on a revolutionary step change in LNG (liquefied natural gas) processing. Prelude will be a 488m long floating facility that will process, liquefy, store and transfer LNG at sea directly above the natural gas field. Could the same floating principle be applied to an alumina refinery moored at a shipping berth close to the bauxite supply?“Going FAR” is, for now, an abstract concept for a Floating Alumina Refinery. The FAR concept potentially reduces the sovereign risk from politically unstable or socially fragile countries by simply moving to another location, protecting the majority of the refinery asset and investment. Going FAR may also provide the opportunity to process small isolated bauxite deposits where it is uneconomical to build a conventional land-based refinery.Bringing this concept to commercial reality will require outside the square thinking, and many critical questions need to be answered. What are the benefits of FAR? How do the economics compare to a traditional onshore refinery? What is the maximum refinery capacity that would be possible as a floating facility? What changes are required in technology and equipment? What are the major changes to safety and environmental risks and how will they be controlled and managed? Will the FAR design inspire innovations which have application for land-based refineries? This paper will explore the edges of these key questions in search of the future direction for this concept.

Sustaining capital cost / expenditure, abbreviated as sustaining capex or Susex, is a project cost item which is sometimes forgotten and often disliked. To many it is unknown, and it may be (partly) hidden in other operating cost items. The relation — and the difference — with repair and maintenance costs is not always clear and may inadvertently lead to incomplete cost estimates.This paper explores several facets of sustaining capital and aims at clarifying some of these issues for alumina refinery projects.

The reaction behavior and conversion of anatase in the calcification-carbonization process were investigated experimentally in the pure substance systems of Na₂O-Al₂O₃-CaO-TiO₂-H₂O and Na₂O-Al₂O₃-SiO₂-CaO-TiO₂-H₂O. The calcification experiments were conducted respectively in the digestion conditions of diaspore and gibbsite. In calcification process, with temperature increasing, anatase converts to firstly Ca₃TiSi₂(Al₂Si_0.5Ti_0.5)O₁₄ and then CaTiO₃. The increase of sodium aluminate solution concentration and additive TiO₂-SiO₂ ratio can promote this conversion. In carbonization process, CO₂ can’t make CaTiO₃ decompose. CaTiO₃ is the most stable titanate phase in calcification-carbonization process.

The alumina production Bayer process generates 0.7 to 2.0 ton of Bauxite residue (BR) and an average of 1.0 ton of CO₂ per ton of alumina produced. The direct use of exhaust gases to react and reduce the alkalinity of BR may allow a triple gain: improving the storage conditions, opening a range of new applications for BR and sequester from 16 to 102 kg of CO₂ per ton of alumina. This paper shows the first results of a lab scale program to measure the effects of adding different percentages of CaO and MgO before carbonation in order to precipitate the alkalinity on stable compounds. . The inlet and exhaust gases were analyzed. The precipitate after carbonation was evaluated to monitor the amount of stable compounds like dawsonite. The pH of the suspension was monitored during and after the reaction to evaluate the buffer effect.

Much work has been done to recycle iron from bauxite residue via reductive roasting-magnetic separation process, of which more than 85% iron could be recycled as powdered metallic iron, but insufficient efforts have been made to utilize the magnetic separation tailings which were rich in valuable components like aluminum, titanium etc. This study was therefore focused on the alumina recovery from the above-mentioned tailings based on alkali leaching. The effects of the parameters like concentration of leaching agent (NaOH), temperature and time to the extraction of alumina have been investigated. Besides, the influence of the up-stream reductive roasting process and sodium additives on the extraction of alumina have also been studied. The results showed the reductive roasting process was favorable to the extraction of alumina while the sodium additives demonstrated an adverse effect. 82.90% alumina was leached under the optimal conditions from a bauxite residue containing 19.59 wt. % Al₂O₃

Activated Alumina was prepared by directly spray pyrolysis AlCl₃ solution. Batch experiments was performed to choose better reaction device among muffle furnace, propane graphite furnace and tubular resistance furnace. The alumina was characterized by XRD and SEM, and the specific surface area of the prepared alumina was also examined. The experiment results showed that using the tubular resistance furnace by spray pyrolysis AlCl₃ solution could effectively improve the efficiency of the pyrolysis, obviously change the morphology of product and effectively increase the specific surface area of product granule. Compared with the muffle furnace roasting, the product index of spray pyrolysis by using tubular resistance furnace was better, and the products conformed to the requirements of the index of activated alumina. The present paper provided a new way of preparing activated alumina.

The modern family of aluminum alloys has a broad range of applications. However, there are currently few aluminum alloys capable of operating at elevated temperatures while maintaining mechanical properties necessary for high performance automotive and aerospace applications. The high temperature phase equilibrium of Al-REE-X (REE-Rare earth elements and X-traditional alloying elements) is not well understood yet this class of materials shows promise as a lightweight material for stiffness-driven high temperature applications. To date, we have determined RE alloys of Al to be castable across a broad range of temperatures and compositions. In addition, room temperature physical properties of Al-RE alloys appear to be in line with existing aluminum alloys with superior microstructural stability and associated properties at temperatures exceeding 150 °C. The CALPHAD work, microstructure, and compositional data are presented for representative example of Al-RE alloys at room and elevated temperatures.

Aluminium can be produced from the raw material bauxite or by recycling aluminium scrap. When aluminium is being recycled, the material strength is then depending on the alloying and trace elements in the aluminium scrap. This paper aims to investigate the sole effect of the alloying element range of Si, Cu, Mg, Mn and Fe on the mechanical properties of the recycled aluminium alloy EN AB-46000 by producing directional solidified samples with low defect levels.The tensile test results and microscopic investigations reveal clearly that the variations within the alloying element range lead to remarkable differences in the mechanical properties that can be quantified to 73 MPa in tensile strength and 49 MPa in yield strength. The elongation to failure was also affected as a function of the alloying elements which is due to increased levels of intermetallics.

Samples of RS-8009 (Al-Fe-V-Si alloy) prepared by both copper mold chill casting and rapid solidification processing were analyzed using SEM and TEM methods including CBED, selected area electron diffraction (SAD), and electron backscatter diffraction (EBSD). Diffraction analysis revealed a hexagonal phase identified as h-AlFeSi (P6/mmm a=2.45 nm c=1.25 nm) with evidence of a structural relationship to the icosahedral quasicrystalline (QC) phase and a further relationship with the α-Al-Fe-Si phase ( <math display='block'> <mrow> <mi>Im</mi><mover accent='true'> <mn>3</mn> <mo>&#x00AF;</mo> </mover> <mi>&#x0020;</mi><mi>&#x0020;</mi><mi>&#x0020;</mi><mi>a</mi><mo>=</mo><mn>1.26</mn><mi>&#x0020;</mi><mi>n</mi><mi>m</mi> </mrow> </math>$$\operatorname{Im} \bar 3 a = 1.26 nm$$), which is the desirable intermetallic in this system and also a quasicrystal approximant. The analysis confirmed the findings of earlier studies on the system, which suggested the same structural relationships using different methods. As will be shown, both phases form across a wide range of cooling rates and appear to have good thermal stabilities.

A201 alloys and A201 alloys with the addition of 1 wt. % silicon were investigated in the as cast, solution treated and aged conditions. They were investigated by a combination of HRSEM, TEM and EDS. The addition of Si was used to improve the castability of the basic alloy, but it was found that 1 wt. % Si wasn’t enough to improve fluidity and prevent hot tearing. It was found that the addition of Si alloy increased the average grain size after solution treatment at 550°C for 19 hours, and large particles of secondary phases were precipitated at the GBs, also increased the precipitation of Ω phase during aging and, by these means, increased the microhardness of the A201 alloy, despite of coarser grain structure of this alloy. There are two kinds of phases were precipitated during aging at 170°C in both investigated alloys, tetragonal lath-shape θ′ phase and orthorhombic polygonal Ω phases with different of the amount, size and shape of precipitates.

Recent trend in the automotive industry towards lightweighting and downsizing the powertrain components, without compromising the power output, have led to increased engine power density. Increased power density frequently requires these lighter components to operate in conditions of increased temperature and pressure, which is challenging for many aluminum alloys in use today in the powertrain manufacturing. Meeting the challenge requires not only improving high-temperature performance of known alloys or developing new ones, but also developing new advanced techniques to understand the long-term behaviour of the alloys.This in-situ neutron diffraction study evaluated creep strain for individual hkl planes in three automotive alloys under tensile load. The conditions of temperature and pressure were typical of or similar to those experienced by engine heads in service. High temperature plastic and elastic creep properties of three cast aluminum alloys were characterized. The properties determined for these alloys can be used as a benchmark for further development of advanced alloy systems suitable for the engine head application.The in-situ neutron diffraction measurements provided data on d-spacing evolution for various hkl crystallographic planes as a function of temperature, tensile load, and time; thereby revealing hkl-specific evolution of elastic strains. This information contributes to in-depth understanding of why a particular alloy exhibits particular service properties (eg. creep).

The amount of aluminum used in cars by the automotive industry has doubled during the last 10 years. Further growth is expected for the following years. Especially 6xxx aluminum alloys, which have the potential to make cars lighter and more economical. The consequential growing demand for high quality aluminum extrusions could be constrained by the problem of intergranular corrosion. The demand for intergranular corrosion resistant aluminum parts is an important issue for the aluminum industry. TRIMET has evaluated a method, which is used for increasing the corrosion resistance of steel. Shot peening is in use for corrosion prevention of steel parts. The prevention mechanism is the introduction of near surface compressive stresses. Additionally some papers also report positive effects of shot peening on the corrosion resistance of aluminum parts. The paper describes the corrosion testing of surface and heat treated aluminum profiles and presents the results concerning corrosion and applicability.

This paper describes the mechanical characterization and corrosion testing of X608 Al alloy that is being considered for A-pillar covers for heavy-duty truck applications. Recently, PNNL developed a thermo-mechanical process to stamp A-pillar covers at room temperature using this alloy, and the full-size prototype was successfully stamped by a tier-1 supplier. This study was conducted to obtain additional important information related to the newly developed forming process, and to further improve its mechanical properties. The solutionization temperature, pre-strain and paint-bake heat-treatment were found to influence the alloy’s fabricability and mechanical properties. Natural aging effect on the formability was investigated by limiting dome height (LDH) tests. Preliminary corrosion experiments showed that the employed thermo-mechanical treatments did not significantly affect the corrosion behavior of Al X608.

The use of Al-alloys is limited in many applications due to a dramatic deterioration of the corrosion properties with any effort made to increase the strength. The hardening phases, uniformly distributed fine intermetallics, act as electrochemical heterogeneities and promote galvanic interactions, thus causing localized corrosion. The chemical composition, size, number, and distribution of the intermetallics in the matrix govern the corrosion of Al alloys. Refining the size and doping of intermetallics via microalloying have been found effective in improving corrosion resistance of Al alloys. This paper presents an overview of recent research on: 1) understanding the influence of intermetallics on corrosion of Al alloys and methods to optimize corrosion and mechanical properties by controlling intermetallic formation via advanced production and post-production processing techniques, and 2) development of ultra-high strength and corrosion resistant Al alloys by combining suitable alloying additions and alloy production techniques.

In this work evolution of texture components during deformation of AA5754 aluminum alloy sheet under cold rolling is studied by analyzing the evolution of element-rotation-distribution calculated using a rate-dependent crystal plasticity finite element model (CPFEM). The proposed criteria can successfully predict the stability of a given textural component for cold rolling deformation in FCC materials with high stacking fault energy that deform predominantly by slip. Comparison of simulation results with experimental data shows that this approach successfully captures the stable textures reported for cold rolled AA5754 sheets. With the initial texture and the strain path, it is believed that, the method described in this work can be used to predict the final stable textures without any need for expensive crystal plasticity based numerical simulations and this could be of immense help for simulating large strain deformation of macroscale sheet samples exhibiting texture evolution.

Because the 5000-series aluminum alloys are not heat treatable, controlling their microstructures during deformation processing to achieve fine grain sizes is particularly important. Recrystallization during or immediately following hot deformation provides an opportunity for controlling grain size prior to cold deformation. The present study investigates the effects of magnesium concentration, temperature, and deformation rate on re-crystallization during and immediately after hot upsetting. Three aluminum alloys with magnesium concentrations of 0.0, 0.5, and 4.5 wt.% were upset at 400 and 500 °C at fixed strain rates of 0.1, 1.0, and 10.0 s−1. The specimens were rapidly quenched after upsetting to preserve their microstructures. Portions of the specimens were then annealed for 10 min. to allow static recrystallization. Cross-polarized optical microscopy was used to examine the as quenched and annealed specimens. These results are interpreted to explain microstructural evolution during and after hot deformation.

It has been shown that dislocation cells induce intragranular backstresses due to blockage of dislocation passage. The present work focuses on the effects of dislocation induced backstresses on the strain hardening behavior of AA 6063-T6 extruded aluminum alloy during cyclic planar simple shear deformation. Based on the previous work of Brahme et al. [1], the internal stresses are predicted using a close form solution based on the Eshelby’s inclusion problem. Using this approach the inherent dislocation microstructure is modeled as a two-phase material consisting of highly dense harder cell walls and less dense soft cell interiors. The new formulation also takes into account the evolution of dislocation cell structure with plastic deformation. Simulations are performed for simple shear and the predictions are compared against experimentally obtained flow stresses for forward, reverse and subsequent forward in-plane simple shearing.

AA 6082 in as received T651 condition was solution treated and immediately processed via equal channel angular pressing (ECAP) to one and two passes and then naturally aged to 100 days. The processed samples were subjected to quasi and dynamic compression testing using regular universal testing machine and split Hopkinson pressure bar (SHPB), respectively. The testing was conducted at a wide range of strain rates and temperatures and the constitutive Johnson model constants were identified. The effect of cryo temperature versus room temperature during dynamic compression was, also, evaluated in terms of the evolution of microstructure and stress — strain response.

The deformation of large structural cast Al components during quenching is difficult to control and easily out of tolerance. Hot sizing based on creep-forming principle is a new technology to correct inaccuracies of parts in shape and dimensions. Thus, creep-ageing curves of Al-Cu-Mn alloy were conducted under constant tensile stresses ranging 10MPa to 150MPa at 175 °C. The curves at high stresses were sparse, while at low stresses were dense. Then, a hyperbolic sine type constitutive model was achieved with genetic algorithm. The creep-ageing model was derived mathematically, but in good consistence with experimental curves with average relative error of 3.85%.

Al-Ti-B type master alloys are routinely added into aluminum for a grain refinement purpose. However, a consensus about the mechanism of grain refinement is still yet to reach. This work is aiming at further understanding the grain refinement mechanism through studying the stability of TiB2 in pure aluminum and TiB₂/Al₃Ti interfaces in aluminum by transmission electron microscopic (TEM) techniques. It is found that TiB₂ reacts with aluminum forming (Ti,Al)B₂ solid solution and releasing titanium into aluminum. A crystallographic orientation relationship exists between TiB₂ and Al₃Ti in Al-Ti-B alloys. A dynamic nucleation theory for the grain refinement is therefore proposed as a dynamic Al₃Ti layer in the thickness of a couple of atoms existing on the surface of (Ti,Al)B₂ and being responsible for nucleating aluminum grains.

A high Mn content alloy (Al-5Mg-1.4Mn) has been cast under the condition of near-rapid cooling. The microstructures and mechanical properties of the alloy have been evaluated. The results indicated that under the condition of near-rapid cooling, the as-cast microstructures of the alloy consisted of α-Al matrix plus two types of intermetallics: Al₆(Fe,Mn) and Mg₂Si. The intermetallics were small and uniformly distributed. During hot and cold rolling, they broke into particles. The particle structures were dense, but small. Some cavities were observed among the Al₆(Fe,Mn) fragments. Both the hot band and cold rolled sheet demonstrated excellent mechanical properties. The significant improvements were achieved by increasing Mn to a higher level, while being cast under the condition of near-rapid cooling.

In the semi-solid high-pressure die casting process, the slurry flows as a solid-liquid two-phase flow. In this study, the effect of the shape of solid particles on their distribution in the slurry was investigated. The solid particles were concentrated in the center of the flow direction in the case of globular-shaped solid particles and high-flow-velocity conditions. Moreover, the concentration ratio of the solid particles increased with an increasing Reynolds number. This phenomenon was explained as follows. The shape of the solid particles affected the viscosity of the slurry; the viscosity was decreased by the globular particles. The solid particles applied Saffman force, which is generated by the velocity gradient, and moved in the direction away from the solid-liquid interface. The Reynolds number increased with decreasing viscosity or increasing flow velocity; therefore, the Saffman force increased with an increasing Reynolds number.

In order to achieve the mechanical properties of thin-wall components made by high pressure die-casting at a level of 300MPa of yield strength, 420MPa of UTS and 3% of elongation, the composition of the new aluminium alloy and casting parameters have been optimized to reduce casting defects in final components. In the present paper, we describe the alloy development in an industrial scale HPDC system.

Dispersoids of Al₃Zr in Al-Zn-Cu-Mg-Zr alloys are important as they pin grain boundaries and inhibit recrystallization during extrusion and solution heat treatment. A high volume fraction, <math display='block'> <mrow> <msub> <mi>V</mi> <mrow> <mi>A</mi><msub> <mn>l</mn> <mn>3</mn> </msub> <mi>Z</mi><mi>r</mi> </mrow> </msub> </mrow> </math>$${V_{A{l_3}Zr}}$$ , and low dispersoid radius, <math display='block'> <mrow> <msub> <mi>r</mi> <mrow> <mi>A</mi><msub> <mn>l</mn> <mn>3</mn> </msub> <mi>Z</mi><mi>r</mi> </mrow> </msub> </mrow> </math>$${r_{A{l_3}Zr}}$$, give <math display='block'> <mrow> <msub> <mi>V</mi> <mrow> <mi>A</mi><msub> <mi>l</mi> <mn>3</mn> </msub> <mi>Z</mi><mi>r</mi> </mrow> </msub> <mo>/</mo><msub> <mi>r</mi> <mrow> <mi>A</mi><msub> <mi>l</mi> <mn>3</mn> </msub> <mi>Z</mi><mi>r</mi> </mrow> </msub> </mrow> </math>$${V_{A{l_3}Zr}}/{r_{A{l_3}Zr}}$$ values above a critical value which are essential to prevent recrystallization. Precipitation of Al₃Zr during homogenization is studied for different homogenization schedules. A 1D finite difference model based on classical nucleation and growth theory for a multicomponent system is developed to simulate precipitation of Al₃Zr dispersoids. The 1D domain represents one half of a secondary dendrite arm spacing with initial concentration gradients based on Scheil type solidification. Single-step homogenization, slow heating to homogenization temperatures, and two-step homogenization are evaluated for maximum dispersoid number density, maximum volume fraction and minimum dispersoid radius. V/r ratios along the length of the SDAS are compared. Two step homogenization (10hrs at 420°C and more than 10hrs at 475°C) provides better V/r ratios across the grain and provides the maximum recrystallization resistance.

Extrusion machining combines cutting with simultaneous extrusion using an additional constraining tool to induce very large and well-characterized strains. Unlike conventional extrusion, accurate analytical models exist for extrusion machining to calculate values of effective strain, strain rate, and temperature rise as a function of processing parameters. Additionally, the size, shape, and microstructural effects of a secondary shear zone that forms due to friction during extrusion machining have been characterized. This secondary shear zone can be manipulated by varying processing parameters, and can produce a severely inhomogeneous microstructure. This microstructure is also seen during conventional extrusion, as the material on the outside of the extrudate is strained more than the inside material, potentially causing an incomplete recrystallization across the face of the product. The secondary shear effect in the extrusion machining process is utilized to reproduce the inhomogeneous microstructure seen in conventional extrusion in order to better understand the recrystallization behavior during AA7050 extrusion.

Heat-treatable 6xxx series aluminum alloys are attractive to the automotive industry for car body sheet applications, particularly for their low densities. Traditionally, aluminum alloy sheets are manufactured by multi-step rolling of cast slabs and annealing. To reduce the processing steps required to make sheet, large strain extrusion machining (LSEM), a shear-based single-step deformation process, was used to create continuous strip from a 6013 aluminum alloy as a model material system. Microstructure and mechanical property of the strips, subjected to a variety of deformation parameters (strains, strain rates and temperatures), were characterized using optical microscopy and Vickers hardness test.

Spray deposition and continuous extrusion (SD-CE) forming technique is a novel technology that combines spray forming and continuous extrusion. Al-20Si alloy rods with a diameter of 8 mm have been fabricated using SD-CE with different extrusion ratios. The microstructure and mechanical properties of these alloy rods were investigated. Al-20Si alloy can be refined effectively by SD-CE, resulting in a great improvement of mechanical properties. The strength and hardness of these alloys gradually increase with increasing the extrusion ratio. However, the elongation rapidly rises with the extrusion ratio, and then gradually drops as the extrusion ratio reach critical level.

Heat-treatable 6000 series aluminum alloys are the most commonly extruded materials in the world. The precipitation process in these alloys is both complex and well characterized. The earliest clustering stage has been shown to have a large effect on subsequent strengthening precipitation, however little is known about the influence of clustering as a function of composition and processing parameters. The current work examines this influence considering the factors of relative and absolute magnesium and silicon content, and the extent of natural aging. Billets were cast and extruded prior to heat-treatment, and the hardening response was evaluated with hardness, conductivity, and transmission electron microscopy (TEM). This work advances the current understanding of Al-Mg-Si precipitation by correlating the kinetics of age hardening to composition and processing, and may lead to further optimization of 6000 series alloy strength and toughness.

Next generation of aluminum automotive engines will have to operate at temperatures approaching 3000C. Traditional aluminum alloys cannot perform at these temperatures, but aluminum alloys reinforced with nanoparticles can. The synthesis of aluminum-titanium carbide nanocomposites by an in-situ gas-liquid reaction implies methane to be injected into molten aluminum that has been pre-alloyed with titanium. The gas is introduced by means of a rotating sparger-impeller unit into the hot alloy, and under the correct conditions of temperature, gas flow and rotation speed, it reacts preferentially with titanium to form titanium carbide nanoparticles that are well dispersed in the metal matrix. The apparatus design, the multi-physics phenomena and a mechanism proposal for nanoparticles formation is first given. The operation window in which to allocate the parametric analysis is next calculated. Finally, characterization of initial obtained material, its relationship to the processing parameters and guidelines to obtain the nanoparticles is done.

Aluminum quasicrystalline alloys present a novel class of high-strength alloys. They possess a high potential for practical applications in many fields. In this study, we used a quasicrystalline Al-alloy with a copper addition to achieving strengthening by heat treatment. The alloys were prepared by melt-spinning and by casting into a copper mold. In both cases, quasicrystals formed during solidification either as a primary phase or as a part of a quasicrystal-containing eutectic. Afterward, the samples were subjected to T5 heat treatment. The hardness of the melt-spun ribbons increased considerably while the hardness of the gravitationally cast samples did not change noticeably. A detailed investigation using transmission electron microscopy (TEM) showed that different types of precipitates can form. At lower temperatures, binary Al-Cu precipitates formed, in the intermediate region prevailed the quasicrystalline precipitates, while at higher temperatures formed T-precipitates (Al₂₀Mn₃Cu₂). Particular attention was given to determination of their orientation relationships with the Al-rich matrix.

Industrial laser marking of aluminium is becoming a key technology for both metal producers and transformers. Its low maintenance and consumables costs due to the fiber-based laser sources, combined with the speed and precision of the marking process makes it a very attractive tool for such industries. This paper aims to explain how to obtain and maintain good contrast laser identifiers on aluminium at various temperatures. The most important laser parameters (spot size, marking speed and line spacing) are analyzed with regards to their effect on the achievable contrast. The results indicate a significant variation of the optimal parameters for the studied temperature range, i.e. between 25°C and 400°C. The marking system tolerance to the piece positioning is also analysed for two different optical configurations.

The aim of this work is to study the corrosion behavior of Aluminum alloys with silicon and magnesium, by potentiodynamic polarization tests and Electrochemical Impedance Spectroscopy in bioethanol solution (92º – 96º), room temperature and pH 6.5, evaluating the effects of the addition of CH₃COOK and ClO₄Li as supported electrolytes. The results show that adding ClO₄Li to the alcoholic solution induces pitting corrosion, possibly reducing ClO₄- to Cl-. Whereas that, CH₃COOK causes inhibition of the corrosion, and increasing of passivation of the metal, a phenomenon which is attributable to the generation of corrosion products.

A novel Al-ZnS master alloy was developed by in situ reaction of Zn and Na₂S in the Al melt. The results from this study left little doubt that this novel Al-ZnS master alloy was a promising refiner in solidification of hypereutectic Al-Si alloys. It refined the primary silicon to the same extent as that achieved by adding P via Cu-P following the same refinement mechanism. The SEM work confirmed that there were many pre-formed ZnS particles contained in the master alloy. The optical metallography showed that the morphologies of the primary silicon crystals in the solidified Al-22Si alloy were drastically changed from irregular coarse morphology to fine regular particles. The primary silicon crystal was refined from 74 μm to 26, 22, 24 μm by adding 0.05, 0.1, 0.15 wt% ZnS respectively with holding time of 20 min, whereas the average particle size of the primary silicon refined by 200 ppm P reduced to 20 μm. With the increase of the holding time up to 1.5 hr, the primary silicon size will increase up to 33 μm in adding 0.1 wt% ZnS.

Effects of Mn addition on microstructure and mechanical properties of as-cast and as-extruded Al-0.15Si-0.2Fe-0.3Cu-0.15Zn-xMn alloys were investigated. After the as-cast billets were kept at 400 °C for 1 h, hot extrusion was carried out with a reduction ratio of 38:1. Mn addition to Al-0.15Si-0.2Fe-0.3Cu-0.15Zn based alloys resulted in the formation of Al₆(Fe,Mn) and Al₁₅(Fe,Mn)₃Si₂ intermetallic compounds and grain refinement. With increasing of addition of Mn, yield strength and ultimate tensile strength was increased and maximum value of yield strength and ultimate strength was 29 and 96 MPa at 2.1 Mn containing alloy, respectively. In the tensile properties at 420 °C, Mn containing intermetallic compounds plays an important role for the improvement of strength and elongation due to grain refinement and suppression of grain growth by the formation of thermally stable Mn containing intermetallic compounds.

The behavior of bubbles was studied in a three-phase mixture of alumina, gas and bath (an investigation in a water model). Thermal and electrical balances were calculated for a low-temperature aluminium cell with bipolar electrodes. The decomposition of alumina dissolved in the AlF₃-KF-Al₂O₃ bath is considered as the basis of cell operation.The cell is designed for the use of vertically oriented bipolar inert anodes. The specific productivity (kg per m3) of such a cell is several times higher than the productivity of a traditional cell. The specific power consumption is about 13 kWh/kg Al.The anode-cathode space (ACS) is filled with a slurry containing more than 30% of alumina.

The primary aluminum producers are continuously raising the line amperage to increase the productivity. Hence, the top heat losses of aluminum electrolysis cells are amplified which leads to higher thermal constraints on the anode cover. A finite element model is developed in order to predict the heat fluxes and temperatures in the top parts of the cell. The model is validated against measurements from the top surfaces of real industrial cells. It considers the cavity above the liquid bath, which is formed during the melting and falling of anode crust during the life of the anode. The heat flux escaping the anode cover increases with the height of the cavity according to the results. Considering the cavity inside the model is mandatory in order to obtain a good correspondence between the measurements and the model predictions.

The AP30 platform reached an important milestone at the Alma Smelter. This latter is the first to operate above 400 kA. Following this success, Rio Tinto Aluminium group launched the AP44 cell development to offer a technology capable of operating above 440 kA. This represents fifty percent more than the original AP30 cell. The technology is expected to deliver world-class performance such as metal production of 3274 kg/day/cell and an energy consumption of approximately 13.23 kWh/kg.Advanced modeling tools and process control system were developed and used to cope with the challenge in designing and operating high amperage cells. Consequently, the AP44 cells integrate the most recent AP technology developments.The industrial piloting is underway at the Alma Smelter on eight dedicated cells to demonstrate the technology at both 405 kA and 440 kA to validate the transition parameters. At the end of September 2016, the new AP44 technology would be available for implementation in existing or new aluminium smelters.

Hydro’s HAL Ultra cell technology development has achieved a major milestone to demonstrate greener aluminium production by operating two different electrolysis cell platforms at energy consumption levels below 12 kWh/kg. Two cells based on HAL300 and HAL4e technology have been rebuild with newly developed solutions to reduce cell resistance, enhance heat conservation and further improve standard operational procedures. Applying Hydro’s portfolio of modelling tools, materials knowledge, process know-how and operational competence enabled the development team to design high performance cell components and methods to reduce process variability. These elements were pre-tested on several cells before merging into full demonstrators. The HAL4e and HAL4e Ultra cell technologies will be installed as part of the Karmøy technology pilot plant.

DUBAL (also known as EGA Jebel Ali Operations) DX+ Technology was selected for the construction of EMAL (also known as EGA Al Taweelah Operations) Potline 3. The 444 DX+ cells installed in Potline 3 successfully started-up over 9 months from September 2013. Amperage was increased from 440 kA to 444 kA during the start-up and reached 450 kA two months after completing start-up. As required in the Technology Licence Agreement, a Performance Test of the DX+ Technology was carried out by the DUBAL Technology team on a group of 30 adjacent cells over a period of 28 days in November 2014. The key performance indicators achieved by the test cells and by the rest of the potline exceeded the guaranteed performance criteria. In this paper, detailed analysis of the results of DX+ cell operation at 450 kA and a description of the Technology Performance Test are given.

In the last decade, the cell productivity has been dramatically improved as demonstrated on AP30 technology with amperage increase from 350 kA to 440 kA. At the same time, available liquid bath volume in the pots has been drastically reduced by the anode surface enlargement. In order to achieve such impressive step change, management of alumina dissolution is a key point to optimize both cell design and process control algorithms.Rio Tinto Aluminium and EPFL have developed a 3D model taking into account the cell geometry, metal and bath velocities, alumina granulometry and thermal balance. It enables to optimize number, location and feeding rate of point feeding systems.In this paper, we present the model development, the latest progress and the model validation with measurements performed on AP technology pots.

In order to understand the impact of anode change on heat transfer and magnetohydrodynamic (MHD) flow in aluminum smelting cells, a transient three-dimensional (3D) coupled mathematical model has been developed. The solutions of the mass, momentum and energy conservation equations were simultaneously implemented by the finite volume method with full coupling of the Joule heating and Lorentz force through solving the electrical potential equation. The volume of fluid (VOF) approach was employed to describe the two-phase flow. The phase change of molten electrolyte (bath) as well as molten aluminum (metal) was modeled by an enthalpy — based technique, where the mushy zone is treated as a porous medium with porosity equal to the liquid fraction. The effect of the new anode temperature on recovery time was also analyzed. A reasonable agreement between the test data and simulated results is obtained. The results indicate that the temperature of the bath under cold anodes first decreases reaching the minimal value and rises under the effect of increasing Joule heating, and finally returns to steady state. The colder bath decays the velocity, and the around ledge becomes thicker.

Mathematical models have become essential for the design of modern, efficient high-amperage reduction cells, but the models are only one part of the cell design process. Since many of the inputs to a thermal balance model are difficult to evaluate, a process for validation of the predictions of the model is essential.The validation process typically includes two sources of feedback: from operational pots and from post-mortem examination. Measurements of temperatures, heat fluxes and ledge shape must be made on the operating pots. In addition, prototype pots are shut down for post-mortem examination (cell autopsy) which is the only way to evaluate transformations of materials.The transformation of materials during operation is one of the major reasons why model predictions do not match real operation. The risk of using unvalidated models to carry on design work is highlighted through the presentation of a real example from the past.

Proper understanding of the sideledge in aluminium electrolysis cells is important for making the right decisions in cell design and operation. A conceptual model for the lower ledge, based on the existence of an alumina-saturated bath film stemming from sludge, was published earlier. In the current study, the rates of freezing and melting were studied in a simplified model, without taking diffusion across the film into account. By considering the heat- and mass balances during freezing or melting of sideledge, it turned out that the system will approach thermal balance as the film creeps upwards along the ledge. As a result, melting and freezing are rapid close to the bottom of the film, while ledge near the top of the metal changes only slowly depending on the thickness of the film. Melting or freezing may also be initiated by variation in the concentration of aluminium fluoride in the sludge.

A CFD based numerical model for a realistic generic aluminium reduction cell is published. The model takes into account the complex physics of the magnetohydrodynamic problem in the reduction cell by coupling of two immiscible liquid flows, Lorenz force, flow turbulence and large length-scale with complex cell geometry. The current model extends the traditional box model with a detailed cell geometry including side and bottom ledge profiles and all channels (side, central and cross channels). The simulation tests of this model focus on quantifying the sensitivity of different side and bottom ledge profiles, and the cross channel by comparing the predicted metal flow and metal pad heaving. In addition, the model dependencies upon model assumptions of open/closed channel top are discussed. Finally a physical mechanism for the metal pad heaving that seems to explain at least parts of this complex phenomenon are highlighted.

The better understanding of anode bubble behavior during aluminum electrolysis can help not only improve the “noise control”, but also optimize the operational conditions and parameters. The effects of current density, electrolysis temperature and alumina content on bubble behavior were investigated in this study. The voltage signals were analyzed using Fast Fourier transform to study variations in the frequency responses. This study demonstrates that the size of bubbles become smaller with the increase of current density, electrolysis temperature and alumina content. Bubble release frequency increases with increase of current density, electrolysis temperature and alumina content. The amplitude of cell voltage fluctuations decrease from 80 mV to 15 mV as current density increases from 0.4 to 1.0 A/cm2. The domain frequency increases from 0.1 Hz to 0.2 Hz as temperature increases from 950 °C to 970 °C.

This paper report on an acid leaching method to eliminate lithium from aluminium electrolyte system (CR=3). Both deionized water and HNO₃ were used. Also, the effects of the influences of acid concentration, leaching time, liquid-to-solid (L/S) ratio and temperature on the leaching efficiency were investigated. The results show that the effect of HNO₃ on leaching efficiency is much better than deionized water. The leaching efficiency increases with the increase of HNO₃ concentration, leaching time, L/S ratio and temperature. The optimized conditions for removing lithium for aluminium bath are presented as following: the concentration of HNO₃ is 14 mol/L; the leaching time is 120 min; the L/S ratio is 12.5 cm3g−1 and the leaching temperature is 80 °C, the leaching efficiency is up to 93.21 %. This new process generates Na₃AlF₆ and LiNO₃ as products from aluminium electrolyte containing lithium salts.

The solidification of the electrolytic bath in Hall-Héroult cells includes the formation of a sideledge, crust as well as temporary freeze layers around alumina agglomerates, new anodes or any cold object inserted into the bath. The structure and the chemical composition of solidified phases depend on the cooling heat flux (or cooling rate), which has a huge impact on the formation of crystals, amorphous phases and pores and on the extent of ion migration. The structure and chemical composition determine the heat loss through the side ledge, and the duration of the existence of a frozen bath around alumina agglomerates or new anodes. The authors of this paper created frozen bath samples under different heat flux conditions in well controlled laboratory environment in order to study the variation of the chemical composition and structure with the cooling rate. The results of this study are presented in this article.

In the Hall-Héroult process, sulfur impurities may not only emit harmful gases but also reduce current efficiency. To better understand this process, the behaviour of sulfur compounds in a cryolite-alumina melt at 1253 K (980 °C) was investigated in a laboratory cell. Sodium sulfate (Na₂SO₄) was added into the molten bath as a sulfur source. Furnace off-gases were passed through a mass spectrometer for qualitative assessment. The stability of sodium sulfate in the cryolite melt was found to depend on the presence of carbon (t_1/2 = 116 min) and carbon-aluminium (t_1/2 = 29 min). It changed dramatically during electrolysis (t_1/2 = 5–8 min). Detected sulfurous gases included SO₂, COS, CS₂, and H₂S.

“Electrochemical short circuiting” describes the loss in current efficiency (CE) due to polyvalent impurity elements that are reduced at the cathode and re-oxidized at the anode. In aluminium cells; the oxidized form of the impurity reacts with dissolved metal near the metal-electrolyte interface, leading to a steeper concentration gradient of dissolved metal. The phenomenon was studied in a model based on the Stefan-Maxwell equations for diffusion in multi component systems and a description of the turbulent diffusion coefficient close to the cathode. The results indicate that the main factors for the slope of CE vs. impurity concentration are the CE in absence of impurities and the change in oxidation state during the reduction, while the diffusion coefficient of the impurity is of minor importance. The numerical results could be represented by an analytical expression that fits well with experimental data.

The concentration of sodium impurity in the aluminum metal pool of operating prebake cells has been shown to be an indicator of cell performance; in general, cells operating with higher current efficiency tend to have a higher sodium content in aluminum. The higher sodium content is due to the increase in sodium containing species in the electrolyte near the bath-metal interface as a consequence of reduced hydrodynamic stirring due to cell MHD forces, anode gases and disrupting cell operations. Correlations between the sodium content in aluminum metal and changes in potline amperage, ac-distance, cell operations, bath chemistry, etc. are discussed for individual cell studies and potlines. Correlations are demonstrated between sodium and other alkali/alkaline metals, (lithium, magnesium and calcium) in aluminum in equilibrium with the cryolite electrolyte.

The electrolyte based on the molten KF-AlF₃-NaF system was investigated for low-temperature aluminium electrolysis in laboratory scale. The current efficiency for aluminium deposition was determined from the amount of aluminium deposited after 4 h at a cathodic current density of 0.85 A•cm-2. The cryolite ratio of molar concentrations of alkali fluorides and aluminium fluoride KR = (N_KF+N_NaF)/N_AlF3, was always retained constant, equal to 1.3 or 1.5. The cryolite-based melts contained 5 and 10 mass% NaF. The temperature was 750 °C for KR=1.3 and 800 °C for KR=1.5. Current efficiencies decreased with increasing content of NaF at cryolite ratios 1.3 and 1.5. The effect of NaF on the cathode process was studied. The contents of sodium and potassium in aluminium based on the KF-NaF-AlF₃ melts were determined as a function of the cathodic current density in the range from 0.5 to 1.2 A•cm-2.

The behaviour of phosphorus and silicon in industrial aluminium cells was measured in QATALUM cells by analysis of bath and metal after addition of AlPO₄, Na₃PO₄ and SiO₂ to the bath. Phosphorus showed a much longer retention time in the bath than silicon. Start-up and anode effects influence the phosphorus and silicon level in the cells. Data trends showed that the amount of phosphorus decreases with increasing temperature. The effect of phosphorus on current efficiency was determined by measuring the metal inventory by copper dilution. The results were compared with current efficiencies estimated from measuring sodium concentration in the produced aluminium. The mass transfer coefficient for silicon was determined to be 3.1 × 10-5 m/s. Mass balance for phosphorus was carried out. It was found that massive amounts of phosphorus entering the cells were transported with the off-gases to the dry scrubbers.

Efficient aluminum smelting operation requires routine but necessary measurement and control of individual cell key parameters. Two of the critical measurement and control parameters are cell temperature and bath chemistry. Bath chemistry of interest for the cryolitic bath is cryolite molar ratio and other components such as CaF₂ and Al₂O₃, and sometimes other additives such as LiF and MgF₂. Determination of these properties requires various analytical equipment and rigorous procedures from tedious sampling to analytical preparation. STARprobe™, an Alcoa proprietary instrument specifically developed and designed for use in aluminum smelters, combines tedious time-consuming sampling procedures and all expensive analytical equipment into one simple portable instrument for potroom operators on potroom floor. Now, the STARprobe™, together with Alcoa proprietary Integrated Pot Control system, becomes an integral technology component in Alcoa smelters for an efficient operation. This paper will update our further development in more accurately measuring alumina concentration, and also addition of property measurement including %CaF₂ and %LiF when needed.

The unsteady part of a cell pseudo-resistance signal called noise is closely related to electro-chemical and magneto-hydrodynamics processes, which is consequently reflected in its non-stationary and transient character, particularly in cases of different disturbances. Therefore wavelet transformation has been introduced in order to examine cell noise simultaneously within time-frequency space, thus enabling a deeper view of the process. A special data acquisition system has been installed on the AP18 Pechiney cell, with the capability of conducting continuous measurements of the cell’s operating data, performing wavelet analysis and evaluating a result using the data obtained from the plant control system. In this paper the analysis and visualization of cell resistance noise, based on continuous wavelet transformation and wavelet packet transformation is proposed, in order to obtain complementary data for more efficient process control. The research results showed promising potential for using the wavelets for the visualization of dynamics and the detections of different disturbances during the aluminum reduction process.

In this paper, The structures and properties changes for single unit cell of alumina, Al-O-C compounds and aluminium carbide were simulated by first principles method. According to theoretical calculation results, alumina carbothermal reductions with coal and charcoal separately used as reductant were carried out under vacuum. The major equations in carbonthermal reduction reduced to three general equations by reaction temperature; the sequence for each substance which formed in reductions was Al4O4C→Al2OC→Al4C3

Current efficiency and energy efficiency are the two most important metrics for assessing the overall performance of a given potline. The influence of alumina quality on these parameters is however poorly understood although likely influential. This interplay is considered here with the construction of multiple regression models from daily data of test groups of 60 cells each for a number of important parameters, pertinent alumina properties from detailed materials characterization of judiciously chosen aluminas and relevant weather parameters for approximately 20 months of effective sampling. The concentration of α-Al₂O₃ in the fines is found to be the single most influential parameter affecting current efficiency and energy efficiency. Models of superheat and noise indicate that this is predominantly due to issues associated with alumina dissolution. Energy efficiency is also correlated with gibbsite content of the fines, likely due to the added energy required for the phase transformation from hydroxide to sesquioxide.

For design and operation of highly efficient aluminium electrolysis cells, knowledge concerning side ledge formation and heat loss through the sides is indispensable. Measurement of ledge profiles and heat fluxes seems to concur with theoretical models only by including a thin film of bath located at the metal-sidewall interface. A transient CFD (Computational Fluid Flow) model including metal circulation is made for these initial studies of how bath film develops. The model includes two liquids (bath and metal) with a bath layer initially sitting at the lowest two cm below the metal, before it proceeds upwards along the side wall by gravitation and surface tension. A parameter study was conducted with surface tension, bath film properties, and metal flow as parameters. It is shown that metal flow has a considerable impact on the film formation rate. The modelled bath film behaviour is in accordance with theoretical considerations.

A pilot test of an aluminum electrolysis was completed in the year of 2014 using inert anodes with the composition NiFe₂O₄-M (M is metal). The volume of the cell is 110 × 66 × 79 cm, the size of inert anode is Φ34 × H22 cm, the total number of inert anodes in the cell are six, the current density of the inert anode is about 0.8 A/cm2, the total electrolysis current fluctuates from 2700 to 4000 A, the electrolysis duration time is 500 hours. The result of the pilot test indicated that the electrolysis process can be run successfully, since above inert anodes have better electric conductivity and corrosive resistance. In addition, none of the six inert anodes cracked during the whole electrolysis process even at the beginning of electrolysis when they were inserted in high temperature electrolyte. The main problem of pilot test is that the content of Fe as impurity is about 1.7% in the product, which is higher than the baseline value of GB/T. In order to remove Fe, related experiments are conducting continually.

In this research, thermodynamic analysis and CASTEP package of the Material Studio program which is based on density functional theory (DFT) formalism were used to study the carbothermic-chlorination (AlCl₃) reaction of Al₂O under vacuum. Thermodynamic calculations indicated that AlCl(g) can be generated by carbothermic-chlorination process at 1760K and 60Pa. The interaction of Al₂O and AlCl₃ with C showed that the chemical adsorption of Al₂O and AlCl₃ did take place on C(001) crystal plane, at the same time, new chemical bond have been formed between Al atom dissociated from Al₂O and Cl atom dissociated from AlCl₃ molecule. The result, after 1ps dynamics simulation, indicated that adsorbed AlCl and CO molecules have been generated in Al₂O-AlCl₃-C system, and they would escape from C surface after a longer period of dynamics simulation time. It means that the reaction of Al₂O and AlCl₃ with C can be carried out under given constraint condition.

Ma’aden Aluminum Company is a joint venture between Ma’aden and Alcoa that operates the largest vertically integrated aluminum complex in the world. This paper focuses on the four (4) Gas Treatment Centers (GTC) that were supplied by Fives to treat gases from the 740,000 pot exhaust TPY aluminium smelter.Ma’aden GTCs are among the largest in the world. The paper reviews the special features of these GTCs, which rely on the TGT-RI scrubbing technology including the use of hairpin ducts to cool potline gases, the YPRIOS dual flow system to minimize potroom emissions and the use of HF gas monitoring equipment at the outlet of each filter compartment to improve the tracking of scrubbing performance. It also presents the performance achieved after two years of operation, the minor corrections that were made to improve the reliability of some components and field improvements that could be reused on future projects.

EMAL commissioned Potline 3 in September 2013, comprising 444 DX+ pots that currently operate at 455 kA. Two large Gas Treatment Centers (GTCs) handle the pot exhaust gas, each one connected to 222 pots and processing 2 million cubic meters of gas every hour. Each GTC is powered by six large Main Exhaust Fans (MEF) through 6.6 kV switchgear, with N-1 design capacity. A major shutdown of three MEFs on the West GTC (5332) was planned on 14 April 2015 to do maintenance work on one section of the 6.6 kV switchgear. This activity lasted for six hours, during which several actions were taken to maintain GTC operation with minimum impact on pot performance and emissions. This paper discusses the actions taken before and during the shutdown, the resulting environmental impact and the lessons learned from this event.

Within the primary aluminium smelter, the spatial requirements and capital cost of the GTC remains a major challenge. Due to its large size the CAPEX is predominantly determined by the mass of steel required. A partial solution to reduce footprint and steel weight is to increase the size of the filter modules while subsequently reducing their quantity. Unfortunately this method of footprint reduction is limited by N-1 baghouse flow conditions — the increased filter velocities with one module out of service.Danieli Corus has addressed this limitation in the new compact GTC design where the low pressure filter cleaning system accommodates longer filter bags. In addition the Danieli Corus compact GTC design is configured with a standby module and exhaust fan thereby avoiding the increased flow associated with N-1 conditions. The new configuration is narrow so it can easily be situated in the courtyard which is often congested with roadways, storage silos and material handling equipment.There are simplistic advantages to the new configuration which result in a decrease of 15% in the steel weight and overall CAPEX reduction of 10% from traditional Danieli Corus design.

RUSAL has been upgrading its aluminium smelters by replacing the scrubbing of the cell off-gases with the “state of the art” alumina dry scrubbers followed by preexisting wet scrubber technology to satisfy today’s environmental regulations for fluoride and sulfur emissions. Today, sixty percent of RUSAL’s smelters are converted to an alumina dry scrubber followed by a wet scrubber and the others are still using a wet-process that captures the fluorides and sulphur emissions with a liquor of soda.While the wet scrubber technology meets recovery efficiency in accord with environmental regulations for the recovery of sulfur compounds for all smelters and it recovers fluorides, the sulfur compounds, mainly as sulfates, end up in costly and monitored landfills.This paper covers RUSAL’s development of a technology for producing a commercial grade sodium sulfate slurry, decreasing salt concentrations, delaying the constructing of landfills, and reducing the environmental impact of wet scrubber technology.

In the natural world, the waste from one species is the food for another species. In the same way, industrial resources can be optimized by looking at how ecological systems operate. This paper describes how the principles of “Industrial Ecology” are integrated into a safe, sustainable solution for hazardous spent potliner (SPL) from the aluminum reduction process. All of the SPL can be detoxified at the source and refined into products which retain the valuable chemicals and minerals in an accessible form. These products have genuine value in energy intensive industries and a Life Cycle Analysis shows a net environmental benefit can be achieved. This approach has been proven over 12 years and market demand is presently greater than supply.

Spent potlining (SPL) is a hazardous waste produced by aluminum smelters. SPL is generated from the internal lining of Al cells, constituted of carbon and refractory bricks and replaced after five to eight years in service. It is classified as a hazardous waste because of its contamination with fluorides and cyanides and its reactivity with water, generating explosive gases. Nowadays, the aluminum industry has made some progress with the SPL issue by recognizing that landfilling is no longer acceptable by most local communities. In 2008, Rio Tinto inaugurated a new plant in Jonquière (Québec) for the treatment of 80 kt of SPL annually, based on the low-caustic leaching and liming process (LCL&L, patent no. US 6,596,252) developed in the early 1990s. This paper describes the advantages of the LCL&L technology, including valorization routes for its by-products and the successful ramp-up of the plant to its nominal capacity reached in 2014.

Reducing sulfur dioxide emissions could be a solution to respect different stakeholders’ expectations, when considering production growth, raw material changes and/or more stringent environmental regulations.Owing to the fact that scrubbing is the selected solution, technologies available today require large equipment to handle and recycle reactants as well as multistage treatments for a barely recoverable by-product, thus resulting in high costs.From 2011 to 2014, Rio Tinto Aluminum division R&D Environmental Group has developed a new patent-pending [1] scrubbing technology named «CHAC», which stands for “Chaux Hydratée Aqua-Catalysée” or Aqua-Catalyzed Hydrated Lime. Tests were conducted at small laboratory scale followed by a larger proof of concept demonstration in a plant environment. The results confirmed the process performance and viability of the new technology. This led to the implementation of a full-scale industrial pilot at the Coke Calciner of the Arvida Smelter, in 2015, to validate the full potential of the CHAC process. The operational and environmental performance was evaluated to confirm the advantages and improvements.

There was an existing cooling tower at Alcoa Aluminerie de Bécancour that had to be replaced by a new improved type to cope with process conditions which had changed since its inception in 1985. The increased volume of Anode Bake Furnace (ABF) fumes, altered pitch burning conditions, anode recipe and the rise in elements such as vanadium and sulfur in the petroleum cokes demanded a larger modern Conditioning Tower design. Danieli Corus (DC) was contracted by the owner’s engineer, Hatch Ltd, to design, supply and deliver the new tower to Aluminerie de Bécancour Incorporated (ABI).During the design and engineering the critical task was to minimize the outage time for tie-in of the new tower to minimize anode production loss and potential fugitive emissions from the ABF. A modular approach was taken in the design whereby the components were fabricated in large sections to minimize the degree of site assembly. Fabrication and pre-assembly of the Conditioning Tower was performed offsite in a manufacturing shop situated in the Republic of China. Inspection and expediting of the components from the shop to the erection site required close cooperation of ABI, Hatch and Danieli Corus BV.Thermal insulating, alignment and pre-commissioning activities were performed over the winter months leading to tie-in in early spring. A very smooth and safely executed tie-in procedure allowed the anode bake furnace and new Conditioning Tower to be brought online within 48 hours.

Alumina is the main raw material for primary aluminium production and the smelters rely on continuous, steady streams of alumina to the pots. At the pot, variations in the alumina quality shall be minimized. Fluoride content, particle size distribution, attrition, segregation, powder flowability and impurities all are important parameters that are affected and partly controlled by the alumina handling and treatments including the alumina silo storage, transport, screening and gas treatment /alumina enrichment systems upstream the pots.The status of alumina handling and research will be reviewed, and new ideas will be presented based on Just-in-time principles, close review of redundancy requirements, storage capacity and how to minimize quality variations all the way from the port to the pot.

The Hyper-Dense Phase System (HDPS™) was developed in the 1980’s by Aluminium Pechiney (AP) to provide its new centre-fed pots with an automatic and cost-effective alumina feeding system. This fluidized conveyor ensures the most stringent technical requirements in terms of alumina quality and reliability are met. Over the years, it has thus contributed to AP Technology’s technical leadership, though it is no longer AP specific. Until recently, it was limited to horizontal conveying, which, in some cases, restricted its potential applications. A major R&D program, which spanned over more than 10 years, was initiated to overcome this limitation and has led to the development of the patented HDPS™ siphon. The successful start-up of the siphons installed at the Rio Tinto (RT) Kitimat smelter concluded this R&D phase. This paper will present this development work and the new potential of the HDPS™ system designed by Alesa Technologies.

In the last thirty years, the majority of greenfield projects, along with some brownfield upgrades, have switched to using pneumatic conveying technology, typically a form of pneumatic technology. Initially, batch delivery technology supplied consistent alumina quality along the potline. With aerated distributed systems, the assumption is still that all pots would receive consistent alumina quality; and typically this has been the case. However, with high amperage pots combined built in long potlines over the last decade, this assumption may no longer be true as variance in both alumina chemistry and physical properties increases with the distance from the Gas Treatment Centre (GTC).This paper summarises the detailed investigation during commissioning of the Pot Feed System (PFS) in the world’s longest potline, Potline 3 in Emirates Aluminium (EMAL, also known as EGA Al Taweelah), an operating subsidiary of Emirates Global Aluminium (EGA). With a nameplate capacity of 550 ktpa, EMAL Potline 3 uses more alumina than most aluminium smelters and pneumatically conveys the alumina up to a distance of 425 m.

In the aluminium electrolysis process, metal is retrieved from the cells using a tapping tube connected to a crucible in which vacuum is made to initiate and maintain molted metal flow. Care must be taken not to contaminate the aluminium with electrolyte during the tapping operation. Continuous improvement of cell performance, including reduction of the metal layer depth, increases the challenges related to the tapping operation. Reduction of the metal flow helps reducing the electrolyte quantity that is tapped, however this increases the duration of the operation and poses scheduling problems.A novel patent pending tapping control method and equipment have been designed and tested in the Alouette smelter, in order to minimize contamination while maintaining or reducing the tapping operation duration. Metal flow control is improved thanks to optimized software. Analysis of the process indicators show that a significant reduction in tapped electrolyte is obtained.The purpose of this paper is not to describe in detail the design and control algorithm of the control unit, which is proprietary, but rather to describe the results on various performance indicators. Some of these results were unexpected and in first appearance, showing downsides to the improved control. Further analysis shows that these apparent downsides are consequences of improved performance.

Crust breakers equipped with bath sensing modules (BSM) were introduced for DX+ cells since the beginning of their operation in 2010. The modules give signals to pot control system (PCS) when a breaker tip touches liquid bath. This paper describes steps in development and optimization of the break and feed control logics from diagnostic of different kinds of breaker failures to problem treatment procedures. Breaker failures can be linked to different problems both equipment and process. Process problems like bath formation on breakers, narrow breaker holes and alumina accumulation cannot be diagnosed automatically but control logic can help by treating them in a semi-automatic mode. It was noticed that distribution of breaker problems is very dependent on breaker locations. The bath sticking on breakers is most typical for middle breakers located in the hottest zone of the cell. On the other hand, narrow holes and alumina accumulation mostly occur in other breakers. The breaking control logic was modified in order to minimize occurrences of these abnormalities.

This update reviews recent results of physical and chemical testing of Si3N4-bonded SiC sidelining material for aluminium electrolysis pots, as well as the corrosion mechanisms of SiC materials under electrolysis conditions. New test methods have been developed to show how these materials perform in a harsh operating environment.The attack by low temperature bath was tested on SiC bricks, also with SiAlON and N-SiC as binder phase. Reverse reactive sintering has become an important manufacturing process for Si3N4-bonded SiC sidelining materialThis review updates [1, 2].

The HF scrubbing capacity of SGA and its HF generating potential have been discussed in recent and past publications. Researchers are still in search for the ideal SGAs with optimum surface area and calcination levels that meet potroom and GTC specifications. It has been shown that pore size distribution is a critical parameter for assessing the suitability of SGAs for GTC operations as the bulk of the accessible surface area for HF scrubbing is attributable to internal porosity. Laboratory scale experiments have been carried out to understand the evolution of the bulk alumina pore network as reaction with humid HF gas proceeds. A reduction in reacted alumina BET surface area and total pore volume is observed with time, after a marginal initial increase. In this paper, we discuss findings on the involvement of pores in the mechanism of HF scrubbing and its possible influence on the fluoride capture kinetics.

Smelter grade alumina (SGA) is the preferred agent for recovering HF from aluminum smelter ventilation gas streams since the alumina also serves as feed to the electrolysis cells, thereby allowing return of the recovered fluoride to the process. This alumina-based recovery loop includes sulfur, since SO₂ evolved from cells is in fact adsorbed on the SGA. This paper explores the competitive adsorption/reaction behavior of SO₂ and HF and on smelter grade alumina as well as the impact of increasing HF exposure on the surface area and porosimetry of SGA.

With the anticipated decrease in available anode raw materials, future anodes are expected to contain higher levels of impurities. Currently the implications of this are being investigated through evaluation of electrochemical performance of anodes and the environmental aspects through gas evolution studies. In this work, four experiments of aluminum smelting have been conducted with anodes of varying sulphur content. The aim of this work was to evaluate the speciation of sulphur in the off-gas. It was found that COS is the main sulphur species in the inert argon furnace atmosphere. The gas composition was further evaluated as a function of set anode voltages in potentiostatic mode. The significance levels of gas composition were evaluated through repeated experiments. As an FTIR multicomponent analyzer was used, less commonly discussed gaseous constituents such as CH4 and HCl were evaluated with respect to given impurity levels in the anode.

Most aluminum companies have initiated voluntary programs for actively reducing propagating, or above 8 volt, anode effect (AE), perfluorocarbon (PFC) emissions and all modern pre-bake smelters have automated reactive methods for terminating anode effects. The goal of these methods is to reduce cumulative anode effect minutes per cell-day. Any PFCs evolved from aluminum electrolysis pots during periods when pot voltage is below eight volts are considered non-anode effect or low voltage (LV) PFC emissions, consistent with current PFC measurement protocols.The inclusion of LV-PFC emissions will additively impact aluminum smelter emission inventories in a manner that depends on a number of location-specific factors such as technology type, operating practices, work practices and AE performance and AE kill strategy. The present work outlines a plant-proven approach for systematically and sustainably reducing both AE- and LV-PFC emissions by optimizing alumina feed control.

Periodic, short-term, extractive sampling with Fourier transform infrared (FTIR) spectrometers is often employed to measure perfluorocarbon (PFC) emissions at aluminum smelters. FTIRs have been routinely deployed by Alcoa to measure tetrafluoromethane (CF₄) and hexafluoroethane (C₂F₆) gases during 2–4 week sampling campaigns at smelters worldwide. In conjunction with plant process logs, the data collected from these short-duration measurements are used to estimate production-normalized plant-specific PFC emission factors. Quantum cascade lasers (QCL) have recently become available for the mid-infrared monitoring of air pollutants. QCL-based, purpose-designed measurement systems could facilitate long-term in-situ stack measurement of CF₄ at one or a few representative dry scrubber exhaust stacks. This paper discusses paired in-plant measurements by FTIR and QCL-based optical systems employed to develop future QCL measurement platforms and optimize the latter for long term stack PFC monitoring.

Primary aluminum production is generating a significant amount of greenhouse gases. CO2 is the dominant compound but during anode effects, perfluorocarbons (PFCs) are released as well. Even when no generalized anode effects are present, intermittent emissions of PFC have been reported in small concentrations but the root causes of these emissions are hardly understood.Measurements were taken at “Aluminerie Alouette” plant using Fourier-transformed Infrared Spectroscopy on individual cells to analyze the evolution of the composition of the gas collected at the duct end. By correlating the variations of the concentration for the emitted gas with cell variables (voltage, intensity, and pseudo-resistivity) and individual anode currents, it was possible to develop a predictive model to quantify the tetrafluoromethane (CF₄) emissions between 10 and 1000 ppb for individual cell emissions.By analyzing the time history of the resulting data and by applying a post treatment process accordingly, it is possible to reduce the number of false predictions and increase precision of the final results.

Most laboratory cells used in the investigation of the alumina reduction process use a single anode. When investigating the initiation of the anode effect an approach with more than one anode might give better results, as the probability of obtaining partial anode effect is higher. Additionally, the design is closer to the industrial, where several anodes are connected in parallel. The system constructed consisted of two anodes in separate electrolyte compartments connected in parallel with a single combined cathode. The results indicate that an anode can go in and out of partial anode effect with little influence on the current, although, kept untreated a full anode effect is likely imminent. The results also show that under certain current and alumina conditions, with only two anodes in parallel, an anode can handle approximately the whole load of a fully passivated anode for a certain time.

PT. Indonesia Asahan Aluminium (Persero) — hereinafter referred to as Inalum is the pioneer of the aluminium smelter in South East Asia region. It was designed by Sumitomo Chemical and started up in February 1982 with the original capacity was 225,000 Tons/year of ingot and 175 kA of amperage.After transforming to be State Owned Company in the end of 2013 year, Inalum had an ambition to continue the optimizing of production to be 300,000 Tons/year of ingots in 2019 year. For the first step, in 2014, Inalum successfully increased production up to 264,474 Tons/year of ingots (more 17.5% than original capacity) by increasing Line current up to 207 kA.The technical improvement objectives are increasing of production capacity by new alumina feeding system, new pot design, anode performance improvement and implementation of water turbine efficiency.

Most recent primary aluminium smelting technologies requires a lot of energy (from 12.8 to 14,5 kWh/t Al). Despite the progress made over the past decades, Alouette has always operated in an energy constrain mode ever since its start-up in 1992. Considering the limited amount of energy available and its vision of sustainability, Alouette has been exploring uncharted territories to make the cell technology (Aluminium Pechiney) more ‘efficient’ by increasing cell productivity while reducing energy consumption. As a result, the smelter has been able to produce more aluminium with its power limitation. Now facing the lower limit of the Anode-Cathode Distance (ACD) of the technology, this has brought additional operational challenges. To overcome this challenging situation, Alouette has developed an assisted anode gauging system on Pot Tending Machines (PTM) coupled with an improved anode setting precision method. This results in an improved cell performance by having a positive impact on Key Performances Indicators (KPI).

The aluminium industry often faces financially conflicting challenges where costs are under increasing pressure but more stringent environmental constraints and customer specifications are imposed. Increasingly, smelters are resorting to treatment of aluminium in crucible technologies to reduce alkali and in particular sodium content in the hot metal prior to delivery to the casthouse. The consequence of this process is a residue or skimming waste that can be costly to process and/or dispose of and typically involves some form of landfill. This paper describes how, faced with a large and growing stockpile of skimming waste, Emirates Aluminium (EMAL, also known as EGA Al Taweelah) used the existing plant to process 100 % of the skimming waste on site with no additional off-site disposal or treatment.

TRIMET Aluminium SE operates four smelters in Western Europe. TRIMET has investigated power modulation and opportunities to provide services to the energy grid for several years. The primary aim has been to stay competitive in a geographic zone with above world-average energy prices as well as having to deal with high volatility in energy prices caused by renewable, non-baseload energy sources.Recent work has seen the implementation of Shell Heat Exchanger (SHE) technology on 12 pots at TRIMET’s plant in Essen, Germany. The heat exchangers are capable of boosting sidewall heat extraction or acting as an insulator depending on active air flow or not. The pots have comfortably maintained heat balance with stable ledge when operated within 150 to 180kA for up to 48h. Greater control of heat loss has also enabled operating voltages to be significantly reduced. By being able to operate significantly above or below the conventional design amperage, the smelter can operate as a ‘Virtual Battery’ to the grid.

When spent anodes are replaced with new anodes, cell resistance is changed. A resistance or voltage modifier should be added to the cell’s base target and then reduced over the load up period to preserve the anode cathode distance. The initial modifier may need to be increased even more to maintain stability. The size and duration of the modifier should supply extra heat to bring the new anode to steady state operating temperature without allowing or causing a significant heat cycle.This paper will explore how to determine the minimum initial modifier and consider reasons for additional modifier increase to maintain stability. How to calculate the total kWh required to bring the new anode to steady state operating temperature will be given. The methods of removing the modifier such that the total kWh will be supplied while seeking to avoid significant heat cycling will be discussed.

Maaden aluminum smelter in Saudi Arabia commenced operation in 12–12–12. The reduction consists of 2 potlines with 360 cell per line (720 cells) at operating amperage of 376 kA.On April 30th, power was lost due to a standby rectifier failure during preventive maintenance. This incident lead to operation of Potline 2 at low load (fluctuated from 290 vs 376 kA) for frequent periods (15 time, 4 – 8 hours each) in order to carry out maintenance in the substation without sacrificing potline control.Ma’aden Rectifiers were assembled using diodes fabricated with a different process (recrystallize structure) that were limiting the rectifiers to operate at a nominal capacity of 105 kA per unit. That makes the cell heat balance control a major challenge for operating at 85 kA below design.This paper covers the successful controlling of the process during the power limitation.

For many years anode dusting has been an issue for aluminum producers. Suspended carbon particles enhance the electrical resistivity of the cryolite bath, setting off a chain reaction of adverse effects for the pot. Many studies have documented the negative influence of carbon dust on current efficiency or bath temperature which are indirect effects of contaminated bath. The dependence of bath resistivity on the amount and on the size of carbon particles is poorly understood though. Tube-type cell experiments were performed to determine the electrical resistivity at 980 °C of standard bath mixtures added with carbon dust. The carbon content ranged from 0.06 to 1.01% in weight with mean particle size of 2 μm. A bath resistivity increase of 70% was measured when comparing the bath with the lowest carbon concentration with the one with the highest. From a bath with a carbon content of 0.06% to a one with 0.16%, the change in resistivity was equal to 13%. Thus the difference in bath resistivity between a benchmark bath with 0.03% carbon content and a bath from a dusty pot will be even larger. These results agree with variations of voltage measured during the ACD squeeze of groups of clean and dusty pots.

The application of individual anode current measurements in the Hall-Héroult cells has been investigated to aid cell monitoring, in addition to the conventional use of cell voltage measurements. Its advantages are significant, especially in high amperage cells, where information from the voltage signal is heavily diluted. One common use of individual anode current measurements is to identify process faults that can cause a significant change in the local current flow paths between anode busbar and the metal pad. While this detection is simple and direct, there are other consequential changes in cell conditions that impact the current distribution in the vicinity of a problem anode. This paper presents a Moving Window Kernel PCA-based method to extract spatial information from the individual anode current signals by incorporating other known process variables.

An appropriate energy balance is critical to achieve high energy efficiency in aluminium electrolysis. However, to satisfy the electricity market, and the market demand for aluminium, a wider operational energy input window is now also required. Unfortunately this would require large scale adjustment of input power and heat dissipation, and is not possible with existing smelter technologies in which heat dissipations are not able to be adjusted widely or regulated. The associated wider cell operational window would also need to be matched to flexible energy supply without causing alumina concentration problems. Side ledge control is also a challenge under such a flexible energy input scenario, and the bath chemistry will be perturbed by dynamic ledge behaviour in conventional cryolitic baths. Thus, observations of the bath behaviour and heat transfer rates in the cell system are necessary in order to understand the limits of long term power adjustment. In this paper an analogue laboratory experiment is proposed, to model the industrial cell system and allow prediction of the effect of large scale power shifts.

In this work we present a single layer neural network based model for bath chemistry variables in the aluminum smelting process. This model is designed to be simulated with real data as if it worked online in parallel with the process. The model is built using a very fast machine learning algorithm, the Extreme Learning Machines, which provides excellent results in regression problems in a very short time. Also we applied statistical analysis for data collection, preprocessing and filtering and for validation we performed several simulations to attest the neural model’s capability to respond to new data. A comparison of this model against linear and traditional nonlinear structures is performed to show how single layer neural networks can be applied on the bath chemistry modeling.

The influence of the materials of cathode carbon block and cathode steel bar, cathode steel bar size and steel bar assembly way on the horizontal current of aluminum liquid and cathode voltage drop was studied using simulation calculation and industrial cell experiment. It is aimed to obtain the technical parameters for decreasing the longitudinal fluctuation of aluminum liquid and reducing the cathode voltage drop, increase the stability of electrolytic cell and realize the cells operated at low voltage and high efficiency. The results of simulation study and industrial cell tests show that it can effectively reduce the liquid aluminum surface waves and the cathode voltage drop through comprehensive optimization of cathode carbon block and steel bar materials, reasonable design of cathode steel bar size and optimizing steel bar assembly ways. The results of 350 kA test cell indicate that the cell voltages are about 3.82–3.85 mV at the current density of 0.73 A/cm2 and the cell voltages fluctuation range is about 6–8 mV. Starting after 180 days, the cathode voltage drop is about 190–220 mV. It realized high current efficiency of production under the condition of low voltage, and reduced the energy consumption for aluminum electrolysis production.

There is a wide range of complex chemical and physical phenomena in the Hall-Héroult process that can be described by mathematical tools. Using the MatLab-Simulink software, a mathematical model has been developed to solve the dynamic status of an aluminium reduction cell. Simulink provides a very powerful graphical user interface for building sub-models as block diagrams. The cell simulator determines various process interactions. The key operating parameters such as alumina concentration, bath temperature, ledge thickness, cell voltage and many others are computed as a function of time. Raw materials and process variations effects are predicted. The model may aid at improving operating strategies that can be implemented in the process control. Applications such as the variation of ledge thickness, specific energy, bath level, AlF3 emissions are presented.

The proportional increase in modern cells’ amperage and anode size while aiming for low energy input has resulted in slowing the process of anode current pick-up rate. Changes in cell thermal and electrical balance due to anode replacement (while a thick layer of frozen electrolyte is formed under a new anode) causes a slower anode current pick-up rate which may require 4 days to reach the target current. Variations in cell conditions following anode replacement result in increasing cell magnetohydrodynamic instability and prolonged presence of low superheat zones. This increases the possibility of forming local abnormalities, such as anode spikes. As part of a continuous enhancement programme, the latest developed individual anode current monitoring system in DUBAL (EGA Jebel Ali) was used to study the benefits of preheating anodes up to 500 °C. The work includes studying the impact of anode pre-heating on current pick-up rate while conducting operations such as anode dressing.

Casting processes of aluminum and especially DC casting processes exist since less than one century. Although this area has a short history the findings in science of casting and the related solidification process of aluminum are key assets for the development of the Al industry and its related fast growth.In his now 35 years lasting industrially based research work Prof. Wolfgang Schneider always focused on the link between sound theoretical understanding and industrial application of scientific findings. This led to various significant contributions to our industry, which is also documented by a number of patents he holds. His development activities span from filtration processes, grain refinement and modification to shape and DC casting processes itself. He published more than 150 papers and received numerous awards for his work such as Light Metals Award in 1990 and 1995 or the Madhu Nilmani Award in 2001.Besides his research he was Professor for Metallurgy at the Technical University of Berlin and gave lectures at RWTH in Aachen. He was also active in related professional networks such as Verein Deutscher Gießereifachleute VDG, Germany, Deutsche Gesellschaft für Materialkunde DGM, Germany, and The Minerals, Metals and Materials Society TMS, U.S.A, where he was the first European President in 2012.

A fully transient 2-D axisymmetric numerical model is used to investigate the effects of liquid metal distribution on the flow field and macrosegregation during DC casting. The model is applied to 500 mm diameter billets of aluminum alloy 7050 cast at 30, 60, and 90 mm/min. The predicted flow fields and composition profiles are compared for three metal feeding systems; one containing only a submerged nozzle, another with the addition of a rigid combo bag, and a final with the addition of a flow diffuser plate. The results from the numerical model show that the flow field and macrosegregation are strongly influenced by the nozzle exit velocity, which is primarily a function of the nozzle to ingot area ratio. For area ratios below 5%, jetting of the nozzle flow down the centerline is predicted, resulting in higher levels of macrosegregation. For area ratios above 5%, jetting of the nozzle flow is not predicted, but oscillations in the flow field at higher casting speeds are. The addition of either a low permeability combo bag or a flow diffuser plate is shown to prevent jetting of the nozzle flow down the centerline and to stabilize the flow field.

Hot-top billet casting technology paved the road to excellence in many primary and secondary aluminum alloys billets cast houses during the last forty years. But the conventional dip-tube and float system kept running many others that couldn’t afford the investments associated with hot-top. This paper intends to shed some light on the future of this peaceful coexistence between hot-top and spout and float billet casting systems.

The initiation mechanism of hot tearing in direct chill casting ingot of 7XXX aluminum alloy was investigated. In 7XXX aluminum alloys (arranged 7055 aluminum alloy), the temperature range between solidus (463°C) and 583 °C was classified into brittle range on mechanical properties and deformation behaviors in semi-solid state. The brittle range of 7XXX aluminum alloys was larger than that of other general aluminum alloys. In 7XXX aluminum alloy, the healing of the crack which initiated from shrinkage cavity was difficult to occur especially in the center of the direct chill casting ingot which was the last solidification part. Therefore, it was considered that the hot tearing in direct chill casting ingot of 7XXX aluminum alloys was easy to initiate the crack in the center of the ingot.

An investigation has been carried out under controlled laboratory conditions designed to approximate the conditions encountered during the early stages of direct chill of Al-4:5% Cu aluminum alloy ingots to study the formation of micro-shrinkage crack. Also, a couple models have been established at the author’s organization to compute convection patterns, thermal fields, interdendritic thermo-metallurgical strain and macrosegregation distributions. Comparisons between the available microstructure taken from the samples and model predictions of convection streams to verify the mechanism of the random sedimentation of free crystals and therefore the micro-shrinkage crack formation were performed where good agreements were obtained. The effects of convection streams of various melt superheats at different locations in the aluminum ingots on the formation micro-shrinkage crack have been studied and analyzed. The mechanism of micro-shrinkage crack formation with different convection streams patterns and superheats were also discussed.

Reduced scrap generation in the value chain is one of the most powerful ways of saving money and energy in the production of rolled products due to less re-melting and metal loss. For the total value chain, geometry deviations are one of the most important reason for scrapped material. Hydro and Hycast has for a long time being working on optimizing the rolling slab geometry to minimize both cut off material in casthouse and scalped off material in the rolling plant. Two decades ago Hydro implemented the flexible mould technology to eliminate the butt swell of rolling slabs.This work describes the implementation of the newest development within rolling slab casting, the Adjustable Flexible Mode (AFM) technology, within Hydro. The resulting geometries, shell zones thickness and the reduction of scrapped material in the total value chain are discussed and quantified by some examples.

Complex solidification processes are often simulated to gain insight into physical phenomena that cannot be experimentally observed. Validation of these models is done through direct comparison to scarce experimental data, in which agreement can be misinterpreted due to uncertainties from the model and experimental measurements. A fully transient numerical model of the direct chill casting process of Al-4.5 wt.% Cu is used to examine the propagation of input uncertainty to outputs of interest. This model solves equations for momentum, temperature, and species conservation. The effect of material property input uncertainties are examined. Probability density functions are calculated based on these input uncertainties for metrics that characterize the ingot macrosegregation and sump depth.

During Direct Chill casting of aluminum alloys technology Wagstaff AirSlip, the primary and secondary cooling causes strong thermal gradients, which leads to the uneven thermal contraction in different sections of the billet. The consequences are manifested appearance of casting defects such as hot tearing and residual stresses. In order to reduce the occurrence of these defects were studied different effects on the formation of hot tearing: changes of technological parameters, chemical composition and microstructure of the alloy EN AW 6060 according to the length and cross section, and the conditions of occurrence and progression of cracks describes the model the ratio of light and heavy metals in the billet. Additional testing of as — cast, cracked and homogenized sample billets in multiple directions cross-section detects that the rapid cooling of the surface of the billet affect the development of fields measured hardness.

The Weibull statistical model with three parameters has been used to assess the repeatability of the high pressure die-castings made by an Al-Mg-Si-Mn alloy. It was found that the round samples had a maximum Weibull modulus, indicating the best repeatability. The machined samples exhibited the second best of Weibull modulus. Among the square samples, the 2 mm thick samples had the lowest and the 5 mm thick samples had the highest Weibull modulus, indicating that the repeatability was worsened for the castings with thinner wall thickness. The microstructural uniformity and porosity levels are critical factors in determining the repeatability of high pressure die-castings.

A significant demand in the use of aluminium alloys, most especially in the automobile industries, has been recorded over the past few decades. This is due to light-weight properties of aluminium alloys which when considered in many industries has a significant increase on the fuel efficiency economy of automobiles. The sustainable production of aluminium from recycling of scraps helps in better energy efficiency, cost effectiveness and reducing carbon footprints. Scraps for recycling can easily be obtained through the post-consumers and post-manufacturers. 100% compatibility of the feedstock from these scraps has been a major challenge in the automotive industry. It is very difficult to get alloys which have compatibility of 100% substitution rate. This study compares and predicts using JMatPro the thermo-physical and mechanical properties of these alloys after mixing and the aftermath effects on their service life

The hot dross pressing technology was first invented for the lead industry in 1896. This early patent was later followed by the hot pressing of zinc slag. The first commercially marketed pressing systems using this 1896 patent as a reference for aluminum, were available in the early 1980’s and manufactured by Pechiney in France and Showa Aluminum in Japan. In the mid 1980’s Anaconda Aluminum also developed a dross press system internally that was used in several of its operations. The first commercially successful dross presses were developed and marketed by Altek International in the early 1990’s. Since that time there have been many commercial manufactures and home made versions of the dross press found around the world. Some have been successful performing the process and some have not been successful. Dross pressing is the most widely used system for hot dross processing used through out the aluminum industry today. The original developing industries of lead and zinc do not widely use this process to recover liquid metals from their slags.

Understanding the performance of aluminum cast house PPE (Personal Protective Equipment) is vitally important for employee safety. The ASTM F955 standard describes the testing of flat sections of PPE fabric and is the basis of this testing. This paper will show that wash cycles affect the performance of some articles of PPE. Vinex fabric was tested as new, washed and dried numerous times with and without fabric softener. Vinex that was heavily washed performed better than new Vinex. The performance of damp Vinex and dry Vinex over a damp undershirt was also tested. A test was developed to evaluate curved fabrics such as the fingers of gloves. The fabrics tested include Vinex, Zirpro wool and cotton.

Grain refinement offers several benefits in aluminum casting applications. It can be performed by inoculation, representing one of the mostly used method at industrial level and is based on the addition of nucleating agents. By using grain refinement method important benefits can be reached, for both cast and wrought aluminum alloys: a fine distribution of the second phases, improved castability, reduction of shrinkage porosity, higher mechanical properties, as well superior fatigue life.In the present paper investigation of the effect of Al-5Ti-1B grain refiner on the microstructure and on the mechanical properties of self-hardening aluminum alloy will be presented and discussed. The appropriate use of grain refiner contributes to improve the mechanical properties, particularly the strength and the ductility of the alloy.

The addition of CrB₂ and Sr into Al-10Si-0.3Mg (wt.%) based alloys not only modify the eutectic Si, but also refine the eutectic grains. CrB₂ particles were found to nucleate the eutectic grains, despite the AlP particles being partly removed or depleted by the interaction between Sr and AlP. Furthermore, P level was found to be one of the dominant factors that affect the nucleation of eutectic grains. CrB₂ addition with a low P level results in a refined eutectic grain size, despite less AlP particles being present to nucleate the eutectic grains. Porosity was found to be mainly located at the boundaries of eutectic grains, highlighting the importance of the refinement of eutectic grains to reduce the size and optimize the distribution of porosity. This investigation demonstrates a novel approach to refine eutectic grains and modify eutectic Si simultaneously into Al-10Si-0.3Mg (wt.%) based alloys, and thereby suggests possible industrial applications.

In recent years the metallurgical community has debated on the influence of floating grains on macrosegregation in Direct-Chill cast rolling-slab ingots. The majority of research to date has been primarily focused on the origin and trajectory of these grains and how to incorporate them into numerical simulations. We propose a novel mixing technique to promote the resuspension of these grains. In order to test this approach, pairs of ingots were cast using the standard DC and mixing techniques. Non-grain refined P1050 ingots were cast in order to illustrate the change in microstructure, while Al4.5Cu ingots were cast to investigate the potential change in macrosegregation. The mixing trials indicate the potential to significantly modify microstructure and macrosegregation in rolling-slab ingots.

Ultrasonic treatment is one of the most efficient and cost effective processes that are developing in today’s technology. It improves mechanical properties, corrosion resistance and homogeneity of the microstructure. In this study, the effect of casting conditions and ultrasonic treatment on the microstructure, hardness and corrosion behaviour of the Al-4.5%Mg alloy was investigated. Alloy ingots were melted in resistance furnace at 750C, and then cast into metallic molds. Pouring temperatures were 750C (superheated) and 639C (liquidus temperature). One sample was poured at 655C and left to solidify in ultrasonic field down to 615C within 2–3 minutes. The microstructure showed a completely coarse dendritic morphology for casting at 750C, and less dendritic morphology with finer grain size when cast at 639C. The finest and most globular aluminum grains and intermetallic particles were obtained by ultrasonic treatment during solidification. Higher hardness values and corrosion resistance were recorded for ultrasonic treated sample in comparison to cast one.

It was found that the efficiency of TiB₂ inoculation to achieve grain refinement was decreased dramatically by Zr addition. In the present work, the mechanism of so-called “poisoning” effect was investigated by advanced analytical microscopy, including Cs-corrected high resolution STEM. For the first time, adsorption of Zr atoms was observed in a mono-layer on {0001} surface of TiB₂. Significantly, atomic resolution STEM Z-contrast imaging revealed that the brightness of atomic columns in the monolayer varied periodically when viewing in <math display='block'> <mrow> <mo stretchy='false'>[</mo><mn>10</mn><mover accent='true'> <mn>1</mn> <mo>&#x00AF;</mo> </mover> <mn>0</mn><mo stretchy='false'>]</mo> </mrow> </math>$$[10\bar 10]$$ direction, indicating that the adsorbed Zr atoms in the layer were ordered in 2D. Deviation by 8 degrees from the well define orientation relationship, (0001)_TiB2 8deg ~ (111)_Al, and <math display='block'> <mrow> <msub> <mrow> <mo stretchy='false'>[</mo><mn>11</mn><mover accent='true'> <mn>2</mn> <mo>&#x00AF;</mo> </mover> <mn>0</mn><mo stretchy='false'>]</mo> </mrow> <mrow> <mi>T</mi><mi>i</mi><mi>B</mi><mn>2</mn> </mrow> </msub> <mo>&#x2225;</mo><msub> <mrow> <mo stretchy='false'>[</mo><mn>0</mn><mover accent='true'> <mn>1</mn> <mo>&#x00AF;</mo> </mover> <mn>1</mn><mo stretchy='false'>]</mo> </mrow> <mrow> <mi>A</mi><mn>1</mn> </mrow> </msub> </mrow> </math>$${[11\bar 20]_{TiB2}}\parallel {[0\bar 11]_{A1}}$$, was observed. The effect of Zr atom adsorption at the interface on mechanism of heterogeneous nucleation was discussed.

Problems with the yield during Mn alloy additions and the occurrence of undissolved Mn rich phases in as cast material containing Ti is the background to the present study of the dissolution of Mn compacts in aluminum melts.The dissolution rate of Mn in liquid aluminum, have been studied in pure Al and in Al-0.12%Ti melts. It was found that Ti additions to the melt decreased the dissolution rate of Mn compacts. It was also shown that the intermediate phases formed at the interface between Mn and liquid aluminum was different after small Ti additions. Moreover, an undissolved Mn briquette, found after casting in a furnace, was examined and the conditions for this to happen have been discussed. The discussion was based on calculations of heat balances during the initial dissolution steps and studies of the transformations occurring within the briquette.

Southwire Company located in Carrollton, Georgia U.S.A. has developed ultrasonic grain refining technology to be used on the Southwire Continuous Rod (SCR) process. Grain refining of aluminum occurs when the crystal size of the solid phases is reduced using chemical, physical, or mechanical means. The ultrasonic grain refining process uses power ultrasound to give rise to non-linear effects such as cavitation, acoustic streaming, and radiation pressure. These non-linear effects are then used to nucleate new grains and break up dendrites during the solidification process of an alloy. The first section of this paper discusses the mechanism of grain formation and refinement using acoustic irradiation. Then the results from the lab trials will be presented. Following this, the benefits of ultrasonic grain refining will be reviewed. Finally the test data from initial scale-up of the ultrasonic grain refining technology at the Southwire Kentucky Plant will be summarized and discussed.

In 2005, Rio Tinto Aluminium started an ambitious program to reduce casting furnace energy consumption in its casthouses. A major reduction step was reached in 2008 by implementing this energy efficiency program in all casthouses, focusing more specifically on the reduction of natural gas consumption (m3 of gas) and specific energy consumption (MJ/t of cast metal). In order to further reduce energy consumption, a new strategy was recently developed and is described in this paper.The main tool of this strategy is a standardized energy performance indicator. This indicator takes into account not only intrinsic furnace operating parameters such as burner efficiency and pressure control but also some specific production characteristics (casthouse layout, temperature of the potroom metal, mass of remelt and mass of alloying ingredients). This indicator allows to adequately compare energy performance between casthouses and to identify best practices and potential improvements. Some industrial applications of this tool are also presented.

Energy makes up a major portion of the controllable costs involved in melting aluminum. While technology such as regenerative burners and multi-chamber furnaces improve energy efficiency, these technologies are expensive and difficult to cost justify given current U.S. natural gas prices. Low cost methods to improve energy efficiency without high capital outlays include process optimization, better maintenance, and furnace settings. The burners on many furnaces are over-fired and poorly controlled. This case study shows the results from a trial made with different fire rates and control schemes on a round-top aluminum melting furnace.

PT. Indonesia Asahan Aluminium (Persero) — hereinafter referred to as Inalum — is the pioneer of the Aluminium Smelter in the South East Asia region. It was designed by Sumitomo Chemical and started up in February 1982. The original capacity was 225,000 Tpy of ingot and 175 kA of amperage reduction cell and 30 Tons of Holding Furnace (HF) capacity in the casting house. Since operation commenced in 1982, only HF-706 had been modified to increase capacity to 36 Tons in 2007. Inalum has now successfully increased furnace capacity to be 40 Tons at HF-704 in October 2014. Improvement of HF-704 has been done through optimized lining design by new materials, an improved heat-up period, and optimized reconstruction time.The improvement is done to support the increasing of production up to 300,000 Tpy in 2019 followed by diversification of Aluminium product (100% of Aluminium Ingot to be 60% of Aluminium Ingot, 30% of Ingot Alloy and 10% of Billet).

In the conventional open flame aluminum casting furnace, a large proportion of the heat generated by the burners is wasted. Proper circulation of the flame’s combustion gases could increase the fraction of heat transferred to the molten metal. In this work, we develop models for the aluminum holding furnace with an aim of optimizing the exhaust gas dynamics, for a more efficient casting process. Using the Finite Volume Method (FVM), fully coupled combustion, heat transfer and fluids dynamics models of the furnace are developed, on the commercial code StarCCM+. The exhaust and burner locations of a typical furnace design are varied to examine the effect of combustion gasses recirculation on the furnace’s first law efficiency. In all models, the non-premixed flame (heat source) is simulated using the Eddy Break Up (EBU), 3-step reaction mechanism. The participating media radiation models are adapted for calculating the radiation heat transfer. Turbulence is modeled using the standard K-epsilon approach. Overall, our results highlight ventilation design considerations for the casting furnace.

This paper covers the results of the first trials of an invention called the Heat Flow Transmitter. A pilot project on a casthouse furnace, using a technology package that includes the Heat Flow Transmitter (HFT) and an air curtain, succeeded in reducing the fuel cost by 33%, and is expected to provide a reduction of carbon emissions of 3,200 tons/year for the casthouse.

Continuous flue gas composition measurements (CO, CO2, O2, H2) were made for four rotary aluminum melt furnace heats, with air/oxy/gas combustion. In many cases, during rotary furnace aluminum melting, the furnace atmosphere becomes increasingly reducing (higher CO and H2, lower O2 and CO2) as the heat progresses. This suggests that aluminum is being oxidized.For these four heats, melting aluminum chips, the measured flue gas CO, CO2, O2 and H2 compositions were utilized, along with C, H and O balance equations, to calculate instantaneous aluminum oxidation rates throughout the heat.Calculated total aluminum oxidation loss ranged from 0.5% to 1.4%. Instantaneous oxidation rates are significantly higher at the end of each heat, corresponding to highest temperature just before tapping. Oxidation rate correlates most strongly with measured flue gas temperature, and does not seem to correlate with % O2 participation or % excess oxidizer from the burner settings.

The energy efficiency of aluminum casting furnaces has been widely reported. However, the conventional energy efficiency analyses are based on the first law of thermodynamics which do not shed adequate light on the processes’ degradation of energy. This just gives a general idea of the furnace’s performance with no reference to possible improvement strategies. In this study, we apply exergy analyses on the aluminum holding and melting furnaces to identify the location and causes of energy degradation. The exergy analyses which are based on a real life furnace conditions highlight the possible locations for technology improvement in a typical cast house. With this established, methods of minimizing the cast house’s exergy losses are assessed.

Ceramic Foam Filter (CFF) ElectroMagnetic (EM) priming systems with square-shaped solenoidal coils are in the process of full-scale industrialization. The demands on the EM fields used in such a system are substantially different from the requirements in the existing EM pumping systems used for liquid metals. The same fundamental physics are applied, i.e. Lorentz forces induced by time varying EM fields; however, in the present case the forces are used to overcome the metal surface energy and force metal into the cold air filled pores of the CFF instead of producing bulk metal movement.The present study discusses the design methodology for square shaped solenoidal coils applied to EM priming using a combination of analytical models, experimental measurements and Finite Element Modeling (FEM) with COMSOL® Multiphysics. The EM force field at different induction coil designs (4 [10.16] and 9 [22.86] inches [cm]) is estimated using FEM and validated against experimental results.

The quality requirements of aluminium products are constantly increasing. It is critical that ingots and billets have the least amount of inclusions in the as-cast product. Filtration is therefore applied to all critical products, ceramic foam filtration (CFF) being the most frequently used technique. In an attempt to increase inclusion capture by CFF the influence of the filter surface chemistry on the filtration efficiency of aluminium alloys has been evaluated. Lab scale filters have been produced with coatings of different chemical composition. The filters have been characterized prior to usage and lab scale casts have been made on a pilot unit in conjunction with two LiMCA for inclusion removal efficiency determination. The paper will present the results from the characterization done and the observations will be discussed.

For good quality billet, metal degasing is important consideration in billet casting. In-line degasser processes help to remove 61 to 66% of the dissolved gas. An improved billet quality can increase extrusion speed. A model of degassing delineates the most important key factors for optimization of product design for high quality, productivity and cost savings using computer simulations. The present work focuses on aluminum direct chilled casting process with gas injected in the mold through a submerged entry nozzle. The effects of gas into the melt and a direct casting magnetic field on the mold fluid flow are studied. Considering the effect of complex flow, injection and heat transfer on the gas bubble size the particles model reveals to be an efficient method to study the degassing phenomenon. However, the model treats the gaseous phase as bubbles with multiple sizes providing a new approach to simulate multiphase flow in continuous casting.

Open mold conveyor casting machines for 8–25kg ingot produce millions of tonnes of remelt ingot every year. However, these ingots suffer from cavity formation. These cavities represent a safety risk during remelting of the ingots due to the possibility of water molten metal explosions if the cavities contain water. The formation mechanisms of these cavities and the effect of alloy, machine design and operational parameters are reviewed. Alternative means of controlling cavities are examined and assessed. The need for a new approach to the problem is highlighted.

Following the initial molten metal quality studies with the MetalVision MV 20/20 Inclusion Analyzer1,2,3,4, presented previously, the need for more robust molten metal treatment equipment and methods was identified. Elimination of fluxing gas and improved filtration techniques were identified as opportunities. New degassing and cleaning technologies such as ultrasonic degassing were tested. This paper presents the results obtained during the initial trials of the ultrasonic SCR Ultra-DTM Degasser5. Some unexpected and surprising results led to further investigation (theory and confirmation tests), which then led to modifications in equipment and further process improvement.

Because aluminum alloy castings are becoming commonplace for critical applications in the automotive and aerospace industries, tight control over the cleanliness of the melt (mitigation of solid particle inclusions) and microstructure must be achieved. In order to control cleanliness, it must first be well defined and measured. Very few techniques exist in industry that can quantitatively measure inclusion levels in-situ. Laser-induced breakdown spectroscopy (LIBS) is presented as a promising technique to quantify solid particles, desired or undesired, in aluminum melts. By performing LIBS with subsequent statistical analysis on liquid Al with varying amounts of TiB₂ particles, calibration curves for B and Ti were generated.

The Aluminum casthouse industry is striving to reduce operating cost, emissions and energy consumption. During the last decades technologies have been launched to able drain-free de-gassers with no use of Chlorine or Chlorine containing salt.Ten years ago, Hycast launched the I-60 SIR unit with a unique siphoning concept, and bottom mounted rotors, allowing for long residual time of the Argon gas. This unit works very well for all flow rates with only two rotors. However, for low flow rates, the unit over-performs and the purpose of this work was to develop a one-rotor siphon inert reactor for low and medium flow rates up to 25 tons per hour. In this work, the basic design and the metallurgical performance of the new SIR I-25 are presented. The removal efficiency of both Hydrogen and inclusions are well within the requirements in the industry.

The hypereutectic Al-Si alloys constitute an important family of alloys because of their excellent wear resistance and low thermal expansion. The current work aimed at studying the effect of soaking treatment on the microstructure and wear behavior of the ultrasonic melt-treated B-390 hypereutectic Al-Si alloy. The results showed that the ultrasonic treatment resulted in uniform distribution and refinement of the primary silicon particles. The soaking treatment resulted in coarsening of the primary silicon particles, with larger particles obtained at longer soaking times. It was also observed that particle coalescence is the primary ripening mechanism for the primary Si, especially in the ultrasonic-treated samples. The wear weight loss changed with the soaking time, with the ultrasonic-treated samples exhibiting more consistent trend with little variations of weight loss at different soaking times. The optimum particle size and distribution is obtained after 5 min. soaking of the conventionally cast samples, and 10 min. for the ultrasonic-treated samples. The worn surface of these conditions is characterized by fine and shallow wear scratches with little particle detachments. Soaking at high temperature leads to reducing the volume fraction of the primary silicon, which consequently, caused high weight losses during wear test.

Aluminum can be easily oxidized at high temperature so that it is difficult to accurately measure the contact angle between the liquid aluminum and Al₂O₃ substrate. In the current study, the measurement of the contact angle between liquid aluminum and Al₂O₃ was performed under Ar gas purging and vacuum conditions. The effect of oxidation on the contact angle between liquid aluminum and Al₂O₃ was investigated. It was observed that aluminum samples in both Ar gas and vacuum condition were oxidized. With further increasing of the temperature, the oxide film gradually reacted with liquid Al and became thinner. The disruption of the thin oxide film caused the mutation of contact angle. The contact angle after mutation was approximately 130° for both cases. The temperature of the turning point was lower under vacuum condition than under Ar gas purging. Under vacuum condition, the final stable contact angle was approximately 63° above 1275 °C.

Ultrasonic cavitation treatment of melt significantly improves the downstream properties and quality of conventional and advanced metallic materials. However, the transfer of this technology to treating large melt volumes has been hindered by a lack of fundamental knowledge, allowing for the ultrasonic processing in the melt flow. In this study, we present the results of experimental validation of an advanced numerical model applied to the acoustic cavitation treatment of liquid aluminum during continuous flow [1]. This was achieved by using a calibrated high-temperature cavitometer. The acoustic spectrum was analyzed at various points across the launder while acoustic pressures were calculated at the frequencies of interest.

PoDFA, widely used in Aluminium cast houses to monitor the concentration of non-metallic inclusions, takes a two-step approach. An initial filtration step increases the particle concentration to a level suitable for the subsequent metallographic investigation of the solidified filter cake, giving information on the amount and nature of the inclusions present. The filtration step is generally done on-site using the PoDFA equipment, but equipment availability or operational difficulties may make it necessary to apply the “Cold PoDFA” procedure. Here the liquid metal is solidified on-site and remelted offline before filtration.Trials with simultaneous Hot PoDFA and Cold PoDFA sampling methods were conducted in different alloy groups to evaluate a) the effect of the Cold PoDFA process on the particle concentration, b) possible changes in the particle nature due to reactions occuring during remelting and c) the influence of the sample cooling rate and the remelting procedure on the PoDFA results.

Non-metallic inclusions have a crucial impact on the quality of aluminum products and their behavior in aluminum melts needs to be understood in order to control them. Oftentimes LiMCA (Liquid Metal Cleanliness Analyzer) is used for the quantification of non-metallic inclusions in the melt. However, since LiMCA provides data with the assumption that all particles are spherical, it is currently impossible to obtain information about the particle shape during an online detection. In this work, aluminum oxide particles were added into an aluminum melt and the inclusions were detected by LiMCA in a demo-scale crucible furnace to observe how particles with different shapes settle. The effect of blocky alumina particles versus thin alumina films on the experimental results was assessed. Analytical modellings were also performed to support the experimental findings. First results on the impact of the shape factor on settling will be presented in this paper.

Crucible induction furnaces are widely used in the aluminum industry, for scrap remelting, metal treatments and casting. The operation principle results in an intense circulation within the furnace, raising a specific question with regard to inclusion dynamics within the melt, reflected by LiMCA measurements at the furnace exit. In an effort to understand the involved phenomena, a hydrodynamic model of an induction furnace was built and complemented by an inclusion module that takes into account the transport of inclusions and the interaction of inclusions with other inclusions (aggregation) or with the crucible walls. A numerical inclusion distribution has been developed that reflects the characteristics of the inclusions present in the melt. The model and results of its application are presented in this paper.

Carbon anodes are made of calcined petroleum coke, recycled carbon materials (butts and anodes), and coal tar pitch. The pitch and coke properties strongly influence the anode quality. High-quality anodes help reduce the energy consumption and environmental emissions during the electrolysis. Methods have been developed to characterize the structural aspects of green and baked anodes, coke-pitch and biocoke-pitch samples from the sessile-drop wettability tests as well as the solid particles of various pitches. These methods are based on the analysis of the images obtained by the optical and scanning electron microscopy techniques. Such data help determine the possible impact on anode quality. Biocokes could serve as alternate raw materials for the partial replacement of petroleum coke, and biocoke samples were prepared from wood by calcination at high temperatures. The wetting of biocokes by pitch was compared with that of petroleum coke. This article presents the results of the characterization work.

High quality anodes require stable raw materials. As the quality of the coke available for production of anodes is changing, future production will rely on cokes that are more isotropic with higher impurity levels than traditional raw materials. The purpose of this work was to improve the understanding of the electrochemical reactivity of different cokes. An-ode:electrolyte wetting is a key factor towards determining electrochemical reactivity, which in turn is affected by anode properties, type of coke and polarization. Pilot anodes were fabricated from single source cokes; one anisotropic coke low in impurities, and one isotropic coke with significantly higher impurity levels. Electrochemical characterisation included chronopotentiometry, cyclic voltammetry and impedance spectroscopy. Wetting properties were studied on both unpolarised and polarised samples by a dedicated wetting apparatus, and indirectly by computed tomography (CT) images of frozen electrolyte films.

Prebaked anodes are made of dry aggregates (coke, recycled butts and rejected green/baked anodes) and the binder pitch. During the mixing process, the wettability of coke by pitch influences the anode paste behavior. Coke particles (-125 +100 μm) were prepared from the same source using two different procedures, and the wettability of cokes by five pitches with different properties was studied using a sessile-drop system at 170ºC. The contact angle results show that the wettability of coke by all pitches follows a similar trend for both cases, but the actual contact angles are different. The cokes and the pitch-coke interfaces were also investigated by the optical and scanning-electron microscopy (SEM) techniques. The image analysis results indicated that the coke particle size distribution depends on the coke preparation, which seems to influence its wettability.

In aluminum industry, it is important to determine the concentration of contaminants present in anode raw materials as rapidly as possible in order to adjust the anode recipe. Sodium, which is an impurity coming largely from anode butts, significantly influences the anode reactivity, and an increase in its concentration increases the anode consumption. A simple and inexpensive method was developed to quantify the sodium content in dry aggregates and anodes without grinding the samples. The method is based on potentiometric principles using a sodium-ion specific electrode. A sample can be analyzed easily within thirty minutes. In order to prevent the rapid degradation of the specific electrode due to experimental conditions, sodium is extracted from samples by electrophoresis prior to the test. The comparison shows that the measured sodium concentrations obtained are similar to the results determined by other test methods.

The properties of calcined petroleum cokes (CPC), used in the anodes in the Hall-Héroult process, are important for the electrochemical behaviour of the anodes. CPCs from different sources vary not only in impurity level but also in microstructure. The trend of available CPCs for anode production indicates an increase in impurities as well as introduction of more isotropic structure. Hence, a better understanding of the behaviour of these materials during electrolysis is essential for optimizing performance. In previous studies, different anode cokes were characterized with respect to anode overpotential and the gas evolution during polarization. Present work focuses on investigation of the interfacial contact between the anode and the molten salt electrolyte. The interfacial contact is important for the electrochemically active area of the porous electrodes, which is different from the geometric area. The focus is on the assessment of electroactive area by measurements of the double layer capacitance, which may be performed during electrolysis. This is a great advantage as the active surface area is known to depend on the polarization. Double layer capacitance is compared for anodes made from very different carbons, and correlated to the overpotential.

Carbon anode, which carries the electricity and provides the carbon necessary for the electrochemical reaction in the electrolytic cell, is an essential component of the primary aluminum production. Energy efficiency and environmental impact are still the key issues in modern aluminum smelters; thus, it is important to control the quality of anodes which strongly influences the energy consumption and the greenhouse gas emissions.As the uniform current distribution appears to be one of the indicators of the anode quality, an experimental system has been developed to measure the current distribution in an anode. This article describes the measurement system used for laboratory anodes and presents the experimental results. The results are also compared with the electrical resistance distribution measurements in the anodes, and the comparison shows good agreement.

The lifetime of an aluminum electrolysis cell is mainly determined by the cathode wear, especially for high amperage cells utilizing graphitized carbon cathodes. The cathode wear mechanisms are, however, complex and still subject to debate. Laboratory tests using inverted cell configurations have previously been used to study commercial cathode materials. Here we report on results obtained in the last 5 years for three different commercial cathode materials. The current density applied in the tests was in the range from 0 to 2 A/cm2 and the rotation speed of the cathode was from 0 to 125 rpm, corresponding to surface velocity 0–19.6 cm/s. The wear mechanisms with respect to current density, limiting current of Al₄C₃ formation and mass transport are discussed. In addition, similarities and differences between wear observed in the laboratory and in industry cells are discussed.

The use of copper inserts in steel collector bars has proven to be an efficient way of improving the Hall-Héroult process. It helps at decreasing the specific energy consumption and improving the cell productivity. A further step is presented for which only copper is used inside the carbon cathode. The feasibility was modeled, tested at laboratory scale and measured in a real cell. The issues, advantages and potential of using 100% copper bars as current collectors inside the carbon blocks are discussed. Modeling results are validated by measurements.

In recent years extensive modeling work has been done to assess if the usage of an irregular cathode surface does or does not increases the MHD cell stability.So far, 2 types of studies have been carried out: full 3D steady-state analysis and 2D shallow layer dynamic analysis. Cell stability being a transient phenomenon, steady-state results are not providing any direct answer to the question.2D shallow layer dynamic analysis can directly answer the question, but unfortunately, irregular cathode surface introduces a third dimension to the flow, so this type of 2D analysis is not the best suited to analyze this type of 3D flow problem.The current work presents a new way to analyze the problem and answer the question. Lateral gravity waves have been simulated in a 3D cell slice model using VOF formulation in OpenFoam. Results obtained for cells using regular and irregular cathode surfaces are compared.

One of the major downsides of the current aluminium production process is the high amount of CO₂ emission. One alternative is to replace the consumable carbon anodes with inert anodes so that oxygen evolves instead of CO₂ and PFC emissions. However, so far a sufficiently inert anode has not been found. Another option is to utilize natural gas through porous anodes. This will decrease CO₂ emission remarkably and also eliminate PFC emissions and anode effect. The porous anode could be made of carbon or it can be inert. However, the as-mentioned problem still exists regarding porous inert anodes. Therefore, at the moment porous carbon anodes seem to be the best practical option. In this study, porous anodes made of different grades of graphite were used for electrolysis experiments. Also, off-gas analysis was performed to get an insight of the ongoing reactions. Our results show that for some types of graphite anodes, methane participates effectively in the anodic reaction.

Investigation of creep behavior of graphite cathode material during aluminum electrolysis in NaF-AlF₃-Al₂O₃ melt were carried out under external 4 MPa pressure with modified rapoport method in a laboratory cell at 965 0C. The effects of cathode current density (i=0 A/cm2, 0.25 A/cm2, 0.50 A/cm2 and 1.0 A/cm2), cryolite ratio (CR=2.34, 3,0 and 4.0) of the cryolitic melts have been investigated. The results demonstrate that increasing current density or CR value can yield enlarged creep strain. The porous structures of the cathode samples were characterized using image analysis, in which the porosity increased with more long and narrow pores appearing after 150 min of electrolysis. The information acquired from this work can be valuable in materials selection and performance improvement of graphite cathodes..

The service life of aluminium reduction cells is limited by cathode wear. A typical wear pattern has been reported as a single or double “W” profile. The uneven cathode wear appears to be controlled by multiple mechanisms such as non-uniform current distribution and carbide dissolution. According to our analyses, formation of aluminium carbide that dissolves into a bath film located on the cathode surface is the dominating mechanism. A dynamic model was developed in order to calculate the local carbide formation as well as cathode wear rate. The present model utilizes a simplified geometry where the metal flow, which is driven by the Lorentz force, results in shear stress and motion of the bath film. The simulations demonstrate how the local current distribution and the mass transfer of aluminium carbide from the cathode surface and into the bath film result in uneven cathode wear profiles.

A composite cathode design with copper inserts could reduce the voltage drop and the horizontal currents in the liquid aluminium, which are responsible for the MHD wave instability and the intense metal circulation. A model incorporating full 3d busbar configuration coupled to the electric current in the liquid is used to study the MHD response when the copper inserts are included in the cathode construction. The results confirm the reduction of electric current component in liquid metal along the collector bar direction, however show the equally important horizontal current in the metal pad in the long direction of the cell due to the current distribution over the outgoing collectors. The time dependent MHD solution is applied to industrial cells incorporating the composite cathode, variable bottom and the full neighbor cell network effect. Examples of two types of cells show the possibility for different scenarios: either decreasing or increasing the MHD stability depending on the overall cell design and operating conditions.

In the early 2000’s, a project to redefine fluewall design was started at the LRF, St Jean de Maurienne, France. The goal of this project was to improve the performance of the fluewall in term of safety, flow and heat distribution, pressure drop and thermo mechanical behavior. The idea came to remove the baffles and, with the help of finite element modelling, a long and difficult development process started with several generations of design that were tested in various smelters across the world. Some pitfalls were found on the road like uneven heat distribution, too fast degassing or fluewall pinching. However, those problems were progressively addressed and baffle less furnaces erected at Tomago in 2008 and Qatalum in 2009 are showing now, after several years of operation, that initial goals are being reached.

Carbon anodes are called the heart of aluminum electrolysis cells and contribute to about 15% of total production cost of aluminum. In case of poor quality, this contribution can reach 25%. The final quality of the carbon anodes and their behavior in the electrolysis cells depend on the conditions during the baking process. In any case, a low-quality green anode cannot be improved during baking. However, the quality of green anodes can deteriorate if the parameters of the baking process are not properly adjusted.The objective of this work is to study the effect of introducing a plateau during baking on the properties of carbon anodes. The samples were characterized before and after baking by measuring their density and electrical resistivity which are important in defining the final quality of the anode samples. The experimental data were also analysed using the artificial neural network model (ANN). The article will present the results of this study.

A soft-sensor model developed from historical carbon plant data and multivariate statistical methods was proposed in past work to obtain quick predictions of individual anode properties right after baking for quality control purposes. It could only be used for anodes baked at the coldest and hottest positions within the furnace due to the core sampling method used at the partner plant and the way the data is collected. To complement the soft-sensor, this work proposes a strategy to account for the thermal history of anodes baked at any position. It is shown that combining categorical variables for pit and baking positions and routinely available firing equipment data is sufficient to predict the temperature history of anodes baked in different positions, which is typically not measured during normal operation. The model will be integrated to the soft-sensor in the future to account for the impact of baking position on anode properties.

In aluminium production, the performance of the anode assembly is a great concern which is mainly associated with its energy consumption, aging, fabrication and cleaning process. Therefore, a better understanding of its in situ behavior is required to efficiently optimize the process using an improved design. In this work, electrical and thermal data have been measured on anode assemblies using a specific in situ instrumentation during a complete anode cycle at Aluminerie Alouette Inc. With the help of a numerical model developed on the ANSYS Workbench platform, the evolution of the electrical resistance and current distribution along the anode assembly components has been determined over a complete anode cycle. Considering these information, an innovative prototype has been designed and gave promising results at the laboratory scale. The next step is to confirm the performance of this new design in a real cell.

Anode assemblies suffer a significant electrical contact resistance across the cast iron to carbon interface that is inversely dependent on contact pressure and area. Industry efforts have incrementally reduced this electrical resistance by increasing stub diameter, changing iron chemistry and by improving the stubhole shape.The additional use of multiple steel conductors to bridge across the cast iron to carbon interface provides a means to further reduce the electrical resistance. The function of the conductors is independent of the iron to carbon contact pressure, the stub temperature, iron chemistry, and the stubhole shape. The steel conductors are tightly driven into the carbon anode at one end, with the other end bonded into the cast iron.This paper includes in-pot performance testing results which demonstrate the reduced resistance when using stubhole conductors.

Based on the recent upgrade of the existing anode baking furnace no.1 at Egyptalum, Egypt, the paper will demonstrate the improvements that have been achieved by introducing a new and flexible firing and control system on the baking furnace.The focus was set on a dual fuel firing capability and strategy. The firing system is able to operate either on heavy fuel oil or natural gas on the same system. This feature provides maximum flexibility in the daily challenges of the primary energy market to carbon operations. The paper covers the most relevant parameters of the baking process such as combustion performance, temperature regime, baking level, anode quality and consistency, fuel efficiency and the resulting operation flexibility.

Electrical resistivity is one of the key performance parameters of prebaked anodes. Nowadays, the desire to lower anodic incidents and energy consumption can be attained with a state-of-the-art equipment, which is able to accurately measure the electrical resistance of baked anodes on a production line. Anode/carbon plant improvements will be achieved by a better understanding of the impact of the physical properties of the anode in the cell and the fundamental causes of anode defects. This paper presents an overview of the milestones and challenges of implementing such a device into a mature carbon plant and ensuring acceptance from the production (or operation) team. This innovative device will provide opportunities for smelters to improve their cost-effectiveness by reducing process variation and maximizing anode performance in the cell.

Over the last decade, metal production at DUBAL (also known as EGA Jebel Ali) has increased by 358 000 tonnes, through expansion and capacity creep, resulting in the plant’s anode requirement exceeding the Paste Plant capacity. This increase was supported by anode procurement at premium from the global market. To handle these challenges, DUBAL Paste Plant embarked on a journey towards best-in-class operational effectiveness where paste plant throughput was increased and the conventional maintenance systems were revolutionized. The Plant throughput was increased by 9 % above the design capacity and an improvement of 19 % in Total Effective Equipment Performance (TEEP) was achieved, while continuously improving anode quality. As a result, anode procurement was reduced by more than 85 000 anodes per year, reducing dependency of the smelter on purchased anodes and resulting in significant financial benefits.

MA’ADEN, Saudi Arabia’s government owned mineral resources company, and ALCOA, the world’s leading producer of Aluminium, signed a JV (Joint Venture) agreement in December 2009 to develop the largest integrated Aluminium complex in Saudi Arabia on Arabian Gulf coast at Ras Al-Khair industrial zone.Aluminium smelter capacity of MA’ADEN is 740,000 mtpy of hot metal. With local bauxite mining capacity of 4,000,000 mtpy, alumina refinery capacity of 2,000,000 mtpy and rolling mill capacity 380,000 mtpy, MA’ADEN is the world largest integrated Aluminium complex.The carbon anode manufacturing of this complex consists of two 40 tph green anode plants using Rhodax dry mix preparation and intensive mixing cascade technologies and one liquid pitch terminal with 2x6,000 tonnes storage tanks.This paper covers the design concept of the green anode plants and its operation performance since its inception in 2011 with the focus being on the commissioning and start-up management as well as the technical challenges encountered.

Within Bechtel Mining & Metals (Bechtel), the Aluminium Center of Excellence tests and validates design alternatives and operational variants to optimize equipment and facility design. In a recent study for a cathode re-lining facility for Ma’aden Aluminium (Ma’aden), a discrete event model (DEM) was developed to demonstrate the achievable potshell throughput by matching planned resources and the expected operating schedule.The initial number of stalls and stations in the cathode re-lining facility layout was based on static calculations. The DEM provided the tools required to optimize the design resulting in capital expense (CAPEX) savings. Also, the need for extra potshells, spent pot lining (SPL) scrap containers and mobile equipment container delivery units was addressed. The model covered the full cycle of re-lining a potshell: from stopping the cell, cathode cooling time in the potroom, delivery of the potshell to the re-lining facility, undergoing all the steps of cooling, de-lining, shell repair, re-lining and return to potroom. Various operational patterns were tested; including the timing of cell shutdown, cathode cooling location (potroom or lining facility), potshell movement by transfer gantry and cranes, shift durations, operating schedule and other variables.

The modern requirements for lightweight products force one to make constructional materials with brand new properties. One such group of materials that meets these demands are fiber-reinforced metal-matrix composites. A combination of light and good deformable aluminum with a strong fiber allows obtaining high strength, ductility and corrosion resistance in a single product. In the present article the application of energy-efficient technology of twin-roll casting for production of flat aluminum-matrix composites reinforced with carbon and glass fiber is observed. The corresponding experimental study on the manufacturing of thin strips of pure aluminum reinforced with carbon and glass fiber rovings was carried out using a 370 mm roll diameter labor twin-roll caster. The obtained flat products of 3…5 mm thickness were analyzed by means of mechanical tests and electron microscopy. The results have shown a great potential of the twin-roll casting for production of high-strength fiber-reinforced aluminum strips.

The main limitation in industrial twin-roll casting of aluminum with respect to productivity is heat removal from the melt in order to ensure complete solidification before the material leaves the roll gap. For this purpose a heat loss analysis was performed along the process chain starting at the melting furnace. By reducing heat losses in the launder system, filter box, and the ceramic nozzle it was shown that melt temperature can be decreased considerably at all stages. This leads to savings of energy and CO₂ emissions. Furthermore, lower headbox temperatures reduce the heat to be diverted in the cooling system. This advantage can be directly converted into an increased strip speed. For example for alloy AA3003, a reduction in headbox temperature by 12 °C did increase the productivity by up to 9% in a full campaign. The findings are now consequently used in production.

A new belt casting process has been developed to cast aluminum-aluminum alloy compound strips. The first strip is cast by a conventional glass fiber belt caster. At a defined distance from the location of the primary solidification, a second casting unit is placed. The local upper surface temperature of the substrate is essential for the forming of a sound metallurgical compound. Numerical simulation models of an existing single strip belt casting process were set up and extended to describe the new double strip casting process. The influence of process parameters (cooling conditions, casting speed) and constructive modifications on the crucial surface temperature of the substrate were investigated. Temperatures measured in the casting strip and in the cooling plates were used to calibrate the simulations. Casting parameters for the new compound casting process as well as the appropriate position of the second casting unit were derived from the simulation results.

Altering operational casting parameters has limited effect on increasing productivity of twin roll cast (TRC) alloys. Further improvement can be accomplished by enhanced heat transfer between shell material and solidifying metal. In that respect copper alloys are commercially utilized in twin roll casting technology. However, it can be easily predicted that microstructure and texture development would be different than those produced by conventional steel shells. Present study investigates crystallographic texture development of as-cast 3105 alloy produced by St/Cu shell pair. Pole figures were employed to characterize the dominant texture. Complimentary studies were performed by macro and microstructural investigations.

Twin roll and twin belt casting method are widely used in the aluminum industry for the production of 3105 alloy for painted products (mainly architectural products). Work hardening curve and final annealing data of 3105 alloy produced from the two casting methods are presented in the paper.Recrystallization annealing was carried out at intermediate cold rolling gauges to study the effect of cold rolling reduction on the recrystallized grain size of 3105 alloy produced by both casting methods. The different casting method results in a different work hardening rate and final annealing response of 3105 alloy.

5XXX aluminum alloys are preferred in automotive applications owing to their high strength, good formability and corrosion resistance. However, these alloys are mostly produced by direct chill casting and there is a limited number of attempt for twin-roll cast (TRC) 5754. With the motivation of the fact that microstructure and properties of sheets produced via TRC are highly influenced by casting parameters and cooling gradient originated throughout the strip thickness, this study investigates the effects of as-cast strip thickness and amount of cold rolling ratio prior to soft annealing on the microstructural evolution as well as mechanical properties of 1 mm thick TRC 5754 sheets. Microstructural characterizations were conducted by optical and SEM examinations on as-cast samples and after intermediate annealing as well as soft annealing performed at final gauge. Mechanical properties of final products processed by different routes were determined by tensile tests and forming limit diagrams.