Experimental investigation of fundamental viscosity, density, and leaching characteristics of sodium-iron-silicate slags

Abstract

Environmental regulations govern the emission of sulfur dioxide for lead smelters. Because of this, some have adopted a hydrometallurgical desulfurization step, which results in a sodium-iron-silicate slag being generated in their furnaces. This slag must be disposed of, but the behavior of impurities from the solidified slag and the physical properties of the molten phase are not well understood. In this work, the leaching characteristics, viscosity, and density of a sodium-iron-silicate system representing a modern secondary lead smelter’s slag have been studied. Slag viscosity increased with decreasing temperature and increasing silica content. At constant silica content, more iron led to a higher viscosity while more sodium led to a lower viscosity. An Arrhenius-type model was produced to predict slag viscosity as a function of composition and temperature. It showed good agreement between predicted and measured values. Density measurements conducted by this study were not precise enough to establish trends with temperature or composition. The slag samples tested in this study formed iron oxides and sodium silicates upon cooling. Satmagan analysis suggested the iron was mostly, but not completely reduced (Fe3+/ΣFe = 0.167). The impurities formed barium silicates and sodium-barium-silicates, lead silicates, and sodium arsenates. In the composition region studied, the samples with more silicon tended to leach less. Further investigation revealed that compositions in the center of the phase diagram (not simply those with less silica) produced extracts with the highest concentration of barium (which exceeded the TCLP regulatory limits). Similarly-high extract concentrations were seen with lead and arsenic from the same high-sodium composition. Composition appears to be a more significant predictor of the extract impurity concentration than temperature in the moderate range of cooling rates. However, when quenching and extra-slow cooling are considered, the cooling rate becomes significant. Barium concentrations were higher for the high-sodium samples which had been cooled more slowly. Quenching, however, can lead to morphological effects which can be detrimental when high-silicon compositions are used. Relating the two phenomena, the compositions which had the lowest viscosities also leached the most. While this would be beneficial for phase separation (as viscosity is in the denominator for terminal velocity of a sphere in a fluid), the leaching characteristics of this slag when solidified would be industrially undesirable. An economic incentive for avoiding this region is that increasing the TCLP pass rate will reduce the expense of hazardous slag disposal. By increasing the TCLP pass rate by 25%, daily profit can be increased by 2.43%.