CIM Bulletin, Vol. 2, No. 2, 2007
I. Wilkomirsky, F. Parada and R. Parra
In most non-ferrous pyrometallurgical processes, the smelting capacity depends strongly on the gas treatment facilities. In the classic pyrometallurgical production of copper from sulphide concentrates, the smelting capacity of different vessels has been increased continuously, both on flash smelting and bath smelting, with a direct increase in the demand for the neutralization of SO2. In virtually all copper smelters, the conventional technology to process the off-gases is to convert the SO2 to sulphuric acid. The increase of the smelting capacity is not necessarily aligned with an equivalent increase in the acid plant’s capacity. The technology for acid production from SO2 is a well-proven technology with less of an optimization window than the smelting technology. In this case, the enlargement of the acid plant or building a new plant might not be economically attractive. Another scenario that could negatively affect increasing the smelting capacity is when the acid markets are far from the smelters. In this case, transport costs can offset the overall operation and the acid production could become an economic burden to the smelter. Additionally, to achieve full capture of SO2, a higher volume of gases with low SO2 content have to be treated.
This situation has led to the creation of a R&D program to analyze the option of reacting CaCO3 with SO2 from gasses of the smelter to form CaSO4:
CaCO3 + SO2(g) + 1/2O2(g) = CaSO4
This reaction is used in the neutralization of the SO2 generated from coal burning plants with a well-established technology and high efficiency, although the SO2 concentration is very low, normally below 0.1 vol.%, while in copper smelters gases range from 2 to 25 vol.%. This makes a difference in the equilibrium conditions and the kinetics. It is not possible to make direct extrapolation of the high efficiency achieved.
This paper presents the results of the steps developed to propose a technology to treat gases with high SO2 content in a fluidized bed system. An initial basic laboratory study obtained the necessary physical chemistry information for the sulphation of limestone with concentrated SO2 gases. This information was the basis for designing and operating a three-stage semi-pilot fluidized bed prototype reactor that was used to develop the required technology. The unreacted core model, controlled by the diffusion of SO2 through the anhydrite layer formed,was validated in the experimental pilot plant. The fractional conversion of spherical particles of limestone to anhydrite are expressed by:
1-3(1-Xi)2/3 + 2(1-Xi) = t/ti
where Xi is the fractional conversion of limestone to anhydrite; t is the time to achieve the conversion Xi; and ti is the time to achieve the complete conversion for a given particle size i.
The semi-pilot runs give results for the conversion of limestone to anhydrite and SO2 capture. A compromise between these two parameters has to be reached ensuring a high efficiency of SO2 capture, which could represent, for some cases, a low conversion of limestone. The parameters that control each of them are the mean reaction time which, from the operational side, are the velocity of the gas, which also depends on the fluidization regime, and the feed rate for the limestone. A complete analysis of the pilot plant threestep continuous fluidized bed reactor is presented where the temperature, SO2 content (2 to 8 vol.%), gas velocity, feed rate, and quality of limestone were tested. The semi-pilot unit was scaled up to a pilot unit built in Codelco’s Caletones smelter in order to validate the results. Virtually, a complete capture of SO2 is possible from gases with low or high SO2 content. The final product, a blend of unreacted CaCO3 and CaSO4, is a non-hazardous product with potential commercial value.