Donald Ziegler

Boundary conditions for lattice Boltzmann simulations

A heuristic interpretation of no-slip boundary conditions for lattice Boltzmann and lattice gas simulations is developed. An improvement is suggested which consists of including the wall nodes in the collision operation.

Stability of metal/electrolyte interface in Hall-Héroult cells: effect of the steady velocity

An instability of the metal-bath interface in Hall-Héroult cells is described. The instability is hydrodynamic rather than magnetohydrodynamic in nature. Above a critical value of the velocity of the fluids (which are assumed to have the same steady velocity), the interface becomes unstable and waves grow; this result is calculated without any specific electromagnetic mechanism. The critical velocity depends on the cell’s flow pattern; for example, a one loop pattern will have a critical velocity different than that of a two loop pattern. The critical velocities calculated are near the velocities measured for aluminum in commercial cells. Also, the form of the waves which are generated is a set of traveling waves whose pattern depends on the pattern of the steady velocity. The variety of observed motions is reminiscent of the variety seen in interface topographical animations generated from measured anode currents. In this analysis, increasing the anode-to-cathode distance decreases the cell’s stability, as indicated by the value of the steady velocity at which flow would become unstable. Two possible technical explanations are developed for this difference from conventional wisdom. Additional study in this area is recommended. Finally, the application of this analysis to an operating cell is illustrated, showing the potential of the method in assessing the stability effects of various operating or design parameters.

Characterization of a Full Scale Prebaked Carbon Anode using X-Ray Computerized Tomography

In the conventional Hall-Heroult electrolysis process, the carbon anode is formed either by pressing or by vibro-compaction. The final properties of an anode are influenced by many parameters such as raw materials properties and manufacturing process. Presently, the aluminium producers have to deal with continuous variation of raw materials properties. To minimize the effects of the raw materials variations on the final product quality, numerical modeling of the forming process is of great interest. However, it is imperative to collect data on real anodes in order to calibrate these models. Some of the most valuable data are the density and porosity distribution of a full-scale baked anode obtained with computed tomography (CT). To test the method, three cored samples of 300 mm in diameter were taken from an industrial anode and scanned with an X-Ray tomograph. Calibration standards were also used to fit the CT scan results with the experimental data.

The moreau-evans hydrodynamic model applied to actual hall-héroult cells

An extension of the Moreau-Evans[1] model for Hall-Héroult cells hydrodynamics is presented. Numerical techniques are used to solve the Moreau-Evans model equations with realistic electromagnetic force fields; the predicted results are compared with those of another model which is the property of Kaiser Aluminum Company and whose results are considered as in fairly good agreement with available measurements (velocity in aluminum, for instance). The main input in this hydrodynamic model,i.e., the electromagnetic force field throughout the two liquids, was previously computed. For a given cell design these data were calculated using the electromagnetic program of Lympany and Evans.[2] For actual cells the forces were deduced from measurements of the magnetic field provided by Kaiser Aluminum Company. As expected, the cryolite flow is found to be governed by the large channels, and to be strongly dependent on the presence of such a channel between the two files of anodes. The use of numerical solution has made possible the analysis of new effects as the interfacial drag and the influence of small channels between anode blocks.

Heat Transfer and Airflow Analysis in the Upper Part of Electrolytic Cells Based on CFD

A CFD model of the top part of an electrolytic cell used in the primary aluminum industry is presented. The model is used to determine average heat transfer coefficients on the main surfaces, under different ventilation rates. Correlations have been developed for the heat transfer coefficient and pressure drop versus pot draft condition. Nonuniformity of the heat transfer coefficient is studied. Finally, the relative importance of natural convection versus forced convection is revealed by the analysis. The knowledge developed in this article is useful for the heat transfer design and analysis of electrolytic cells, which is crucial in this industry.

Effects of Physical Properties of Anode Raw Materials on the Paste Compaction Behavior

The current study investigates the effects of coke particle characteristics and paste formulation on the flowability and the compression behavior of anode pastes. Shape factor and texture of different fractions of cokes were characterized using an image analysis system where the characteristics of each coke were correlated to its vibrated bulk density (VBD). A compression test was designed to study the effects of particle characteristics and paste recipe on the compactability of pastes. The test was applied on four anode pastes, prepared from different coke types, particle size distributions and pitch contents. It was observed that the compression test is significantly sensitive to any changes in raw materials characteristics and formulations. Consequently, the compression test may be used as a tool for evaluating anode quality in relation with material variations.

Air and CO2 Reactivity of Carbon Anode and Its Constituents: An Attempt to Understand Dusting Phenomenon

Carbon anode, used in the Hall-Heroult process, is subject to both air oxidation and carboxy gasification by CO2. These reactions are considered as being the main causes of dusting phenomenon in the electrolysis bath. More precisely, it is believed that CO2 and O2 preferentially attack the baked pitch (pitch coke) resulting in detachment of coke and butt particles in the form of dust. The present work aims at elucidating this phenomenon by studying the air and CO2 reactivities of the prebaked carbon anode and also of its constituents. The air and CO2 reactivities were evaluated using thermo-gravimetric analysis and the standard reactivity test of R&D Carbon, ISO Standard 12981-1. The microstructural features of the samples, i.e. real density, crystallite size and specific surface area, were measured. The reactivity of the anode and its constituents was assessed separately and their effect on dusting phenomenon was discussed.

Characterization of Pre-Baked Carbon Anode Samples Using X-Ray Computed Tomography and Porosity Estimation

Computed tomography has been used in recent years to gather information on carbon anodes which can be used to calibrate numerical models dedicated to simulate the anode forming process. To this end, samples with diameters varying from 50 mm up to 300 mm and cored from an industrial anode have been scanned in a Somatom Sensation 64. A correlation was established between the CT scan results and the apparent density. To validate the correlation, an extended campaign was performed on 50 mm diameter samples cored in 20 different anodes with the advantage of using possibly different raw materials. In addition to the CT scan results, the apparent and real densities have been experimentally measured to estimate the porosity level. Similarly to the apparent density, a correlation between the CT scans results and the porosity has been proposed.

Influence of Mixing Parameters on the Density and Compaction Behavior of Carbon Anodes Used in Aluminum Production

The current study investigates the effects of mixing parameters on the quality of anode paste and green anode used in aluminum electrolysis. Four mixing times and four mixing temperatures were applied to prepare anode pastes using calcined petroleum coke and coal tar pitch, as raw materials. The volume of the pores in the paste was used as the indicator for mixing effectiveness. A compression test was applied to the pastes to study the compactability of these pastes and to investigate the effect of mixing parameters on density of the green samples. It was shown that mixing parameters influence the size, volume and surface area of paste porosity and also its compaction behaviour.

Discrete Element Method Applied to the Vibration Process of Coke Particles

Physical properties of coke particles including particle shape and size distribution have direct effects on their packing density. In the present work, effects of particle shape and size distribution on vibrated bulk density (VBD) of dry coke samples have been investigated. Discrete Element Method (DEM) has also been used to simulate the vibration process. Results showed that the shape and size distribution of particles influence the bulk density of coke and these parameters can be used to describe the packing density of coke particles. In general, mixed samples provide higher VBD than mono-size samples and as the fraction of coarse particles increases vibrated bulk density increases. However, existence of 10 wt.% of fine particles to fill the pores between coarse particles is essential. Simulation results were also reasonably consistent with experimental data. Finally, it is noteworthy that a well-tailored DEM model is capable of predicting the particle rearrangement and density evolution during the vibration process.