Reaction rates for sulfur fixation with iron at 1100 to 1275 K
- PDF / 1,736,785 Bytes
- 10 Pages / 603 x 783 pts Page_size
- 47 Downloads / 157 Views
I.
INTRODUCTION
IRON oxide has long been recognized as an effective reagent for high temperature desulfurization of fuel gases produced from coal. 1,2 In contrast with lime and calcined dolomite, iron oxide used for this purpose is easily regenerated by standard sulfide roasting procedures which yield gases of high SO2 content suitable for sulfur recovery or acid manufacture. Also, with cyclic use of iron, makeup can be provided automatically from the pyrite and other Fe-containing minerals in high-sulfur coals. An alternative to a two-step process o f gasification followed by desulfurization of the gas is a one-step process in which sulfur is fixed with solid or molten iron reagents added directly to the primary coal conversion reactor. Under these conditions, added iron oxides are reduced to metal (for example, sponge iron), and the principal sulfur fixation reactions are: H2S + xFe ~ FexS + H2
[1]
COS + xFe ~ FexS + CO
[2]
where x = 0.99 when "FeS" is in equilibrium with Fe at elevated temperatures) The work presented in this paper was done as part of a study of the technical feasibility of a combined, one-step, gasification and sulfur fixation process. Primary objectives o f the present work were to measure and to characterize the rate o f sulfur fixation with iron under coal gasification conditions. Sulfidation o f iron is encountered as a scale-forming reaction when steel parts are exposed to sulfur-containing gases at high temperatures. Turkdogan and Worrell3'4 showed that the rate-limiting step for this process is the diffusion o f Fe through the FexS layer. McCormick, Dayananda, and Grace5 similarly observed parabolic kinetics for the sulfidation o f wustite and attributed their results to the diffusion of iron through both wustite and the FexS layer to the gassulfide interface. CRAIG B. SHUMAKER, formerly a Graduate Student in the School of Materials Engineering, Purdue University, is now with The Standard Oil Company (Ohio), 3092 Broadway, Cleveland, OH 44115. R. SCHUHMANN, Jr., is Ross Professor of Engineering, School of Materials Engineering, Purdue University, West Lafayette, IN 47907. Manuscript submitted December 16, 1981. METALLURGICAL
TRANSACTIONS B
Most previous experimental studies o f gas-solid reaction rates have involved the exposure of a single particle (for example, plate, pellet, or grain) to a flowing gas environment of constant temperature and composition, with observation o f the extent o f reaction o f the particle as a function of time. This procedure has been applied extensively, for example, to gaseous reduction of iron oxides in work summarized by McKewan,6'7 Themelis and Gauvin,8 Szekely, Evans, and Sohn,9 and others too numerous to cite here. A common objective o f such studies o f gas-solid reaction rates has been to identify rate-limiting steps and to fit rate data to models which correspond to the rate-determining mechanisms. These single particle models can then be used in fluid dynamic models of fixed-, moving-, or fluidbed, multiparticle systems usually requiring co
Data Loading...
