Kinetics of reduction of iron oxide with carbon monoxide and hydrogen

Abstract This investigation was undertaken to determine the rate controlling step in the reduction of iron oxides with hydrogen and carbon monoxide. For the reduction of porous hematite pellets, and bars it was found that the reduction rate is controlled by the counterdiffusion of reactant gas and product gas between the reaction zone and the main gas stream. The reduction specimens were spheres ranging in size from 1.5‐ to 4.4‐cm. diameter and 5.1‐ by 7.6‐cm. bars either 0.69 or 1.24 cm. thick. They were prepared from electrolytic iron powder, partially oxidized in a rotary kiln. The final oxidation of the specimens to hematite was accomplished by firing them at 1,149°C. in an oxidizing atmosphere. The bulk density was 3.5 g./cc. The specific surface area was 0.08 sq.m./g. and the void faction was 0.31. The oxide specimens were reduced in streams of pure hydrogen or carbon monoxide at temperatures between 700° and 1,200°C. The reduction was followed by measuring the sample weight during the reaction. Reduction rates were studied at system pressures of 1 and 2 at. The samples reduced step by step, to magnetite, to wüstite, and finally to iron. A shell of reduction, clearly visible in sections of partially reduced specimens, moved concentrically into the core of the samples. Ahead of the receding interface the specimens were known to be pervious to the reactant gases. This led to the conclusion that the gas composition at the interface was in equilibrium with the oxide phases present. Except for the removal of the last portions of oxygen at the lower reduction temperatures, the course of the reduction followed Fick's first law of diffusion. Values of the diffusion co‐efficients, determined from the experimental data, were in close agreement with values predicted from empirical equations. Varying the total pressure of the system had no effect on the rate of reduction, which is consistent with the equations developed from Fick's law for the diffusion‐controlled reaction under investigation. It is speculated that the retarding of the reduction reaction at the lower reduction temperatures is due to the entrapment of oxide inside shells of iron, which are impervious to the reducing gas. Reduction of the trapped oxide then proceeds by solid state diffusion in a manner proposed by Edström (6).