Loss of current efficiency in aluminum electrolysis cells
- PDF / 627,608 Bytes
- 6 Pages / 603.28 x 783.28 pts Page_size
- 68 Downloads / 267 Views
I.
INTRODUCTION
THE purpose of this paper is twofold: to generate equations which represent operating plant experience and to consider some implications relating to the mechanism of the loss of Faradaic efficiency. The observation on which all subsequent thinking on loss of efficiency has been based was reported by Pearson and Waddington in 1947. t~j They found, to a good approximation, that each 1 pct loss of Faradaic efficiency, in terms of metal production at the cathode, was associated with 2 pct CO in the CO2 gas generated at the anode. They postulated that metal was crossing the cryolite in the interelectrode gap and reacting with the CO2 bubbles; the above relation follows. The hypothesis has been universally accepted, and indeed, measurements of anode gas composition are often used to deduce current efficiency (CE). But how does the metal get across the gap? Pearson and Waddington talked in terms o f metal "fog" (referring to the opacity of cryolite melts when in contact with metal) and presumably thought of dispersed droplets of colloidal aluminum, but today it is known that genuine solution of metals in their molten salts is the rule rather than the exception, and there have been many measurements of the solubility of aluminum in cryolite. They are not easy measurements to make, and there is a lot of scatter; the most recent are those of OdegS_rd, t21 which are internally consistent and are used. Sterten, in a recent paper, [31 has taken a chemical engineering approach to the problem, but his equations are too complex to be of use here, where the first concern is chemistry rather than engineering. The question which a plant operator poses is: If the electrolyte composition is changed, what will happen to the CE? That is to say, the cell design is fixed, the ceils are already operating with some particular electrolyte composition, and their CE is established; what effect will a change of composition have? A number of carefully controlled plant experiments have been made; what is needed is a rational equation to represent them. ERNEST W. DEWING, formerly Principal Scientist, Kingston Laboratory, Alcan International Ltd., Kingston, ON K7L 5L9, Canada, is retired. Manuscript submitted May 29, 1990. METALLURGICAL TRANSACTIONS B
Most of the data on which the present work is based were gathered over many years by, or under the guidance of, the late J.W.J. Burck in the plants of the Alcan group of companies. He described his methods, and gave some of his results, in 1971 .[4] In utilizing that body of data, the present author pays tribute to his memory and is indebted to all others who helped to generate it. Basically, factorial-design experiments were used in a carefully controlled section of a production cell line. Three months were spent getting 32 cells operating at the required targets, and then efficiency was measured over another three months. Great attention was paid to keeping the individual cells in similar condition. Thus, if a composition change resulted in a change of liquidus temperature of the e
Data Loading...
