Response of Cryolite-based Electrolytes and Side-ledges to Flexible Potline Power Shifts at Smelters

Abstract

A technology for flexible potline amperage control is now recognised as a competitive advantage for the aluminium smelting industry, in order to align smelting operations with the power and aluminium metal markets. Appropriate heat balance control is important to achieve high energy efficiency in aluminium electrolysis. To correct the imbalance caused by the amperage adjustment, the Shell Heat Exchanger (SHE) installation on the cell has been recommended by other researchers. Hence, a "Heat Balance Shift" technology is proposed in this study. In the present research, the "Heat Balance Shift" and its influence on the cryolitic bath and side ledge are investigated in a laboratory analogue to an industrial aluminium smelting cell. In this analogue, the heat balance shift is driven by a graphite 'cold finger' heat exchanger (to mimic the SHE) and a corresponding change in the power input from the furnace (to mimic the amperage adjustment). The experimental results of shifting the heat balance between different steady states are reported for the first time, both in the conventional bath system (Cryolite - AlF₃ - CaF₂ - Al₂O₃) and in the Li modified bath system (Cryolite - AlF₃ - CaF₂ - Al₂O₃ - LiF). The effects of such heat balance shifts on bath superheat and frozen ledge are investigated. Both of these key parameters are shown to be a function of the applied heat balance shift and to interact with each other. A low superheat gives rise to much thicker freeze, while, on the other hand, there will be no freeze remaining under a high superheat when a certain maximum heat flux to the heat exchanger is exceeded. Energy balances obtained in the lab 'cell' are compared with published modelling results for a modern industrial cell, and a similar range of heat balances (e.g. superheats) are found in both systems. The microstructure and morphology of the freeze lining are studied by Scanning Electron Microscope (SEM) and Electron Probe X-ray Microanalyses (EPMA). Three layers are detected in the freeze linings, including a closed crystalline layer, an open crystalline layer and a sealing layer. These layers have different morphologies and phases, and contribute differently to each heat balance shift based on the final thermal steady state freeze thickness achieved. It appears that the freeze thickness is dominated by freezing or melting the open crystalline layer. The phase chemistry of the freeze lining is analysed to understand the freeze formation mechanism. X-ray Diffraction (XRD) and Electron Probe X-ray Microanalyses (EPMA) are used to characterise the freeze microstructure and compositions of the phases present in different freeze layers, including the phase identification of cryolite, chiolite, Cacontaining phases and alumina. A freeze formation mechanism is proposed based on the microstructural investigations and also thermodynamic predictions by FactSage. It is found that the liquid entrapped in the open crystalline layer becomes very rich in rejected solutes compared with the pure primary phase of the solidified crystals and the bulk composition of the molten bath. Therefore, this open crystalline layer is rapidly removed by superheat increase in the bulk bath, and also dominates freezing of the lining at low superheats. Li₂CO₃ is added to the conventional bath system as a bath modifier to produce the Cryolite - AlF₃ - CaF₂ - Al₂O₃ - LiF bath system. By comparing with results of those in the conventional bath system, it is evident that the Li modified freeze lining is more sensitive to similar heat balance shifts, i.e. higher melting/freezing rate caused by the superheat change. An industrial side ledge sample is also observed at the macro and microscopic levels and shows a similar microstructure to the laboratory freeze linings. It appears that the experimental freeze lining investigations here can provide the smelters with insights into the side ledge maintenance during heat balance shifts.