The kinetics of the reduction of ferrous oxide with hydrogen in a steady state fluidized bed have been correlated with a rate equation based on n reaction controlled at the oxide-metal interface and on a differential material balance which assumes that a close approach to piston flow prevails. The correlation gives an enthalpy of activation of 29,500 b.t.u./lb. mole for the reaction, which is in good agreement with McKewan's reported value of 27,500 b.t.u./lb. mole. The relationships developed are used to predict the effect of the important independent variables on the over-all kinetics of the system.

No abstract

The kinetics of reduction of ferrous oxide with hydrogen were studied in a 1%-in. diameter isothermal batch fluidized reactor. The effects of mass velocity and pressure on the reduction path were investigated over the ranges 5.3 to 19.2 moles/sq.ft.hr. and 1.3 to 5.4 atm. at 1,OOO" and 1,300OF. The results are correlated by means of instantaneous oxygen balances on the assumptions that there are no oxygen concentration gradients in the solid phase, that the distribution of residence times for the gas flow is represented closely by assuming complete mixing, and that the reduction rate is controlled a t the oxide-metal interface. The rate of reduction increased with increasing pressure and mass velocity. An excellent correlation was obtained for the program a t 1,OOO'F. At 1,300OF. sintering and grain growth of the product Ioyer complicated the reduction process, and a correlation was not obtained. Application of the results to the design of commercial fluidized-reduction processes is discussed qualitatively.

4) should give the activation energy, E, that was assumed in evaluating the quantity, G, if a correct value of activation was chosen for calculating G.For the data from tests with the 31-cm. cylinder, Arrhenius plots of Equation 1 using bath temperatures are shoivn in Figure 7. T h e slope of the line passed through the points corresponds to an activation of 48 kcal. per mole. If Equation 4 is used with E = 28, the points in the Arrhenius plot d o fit a line corresponding to an activation energy of 28 kcal. per mole as shown in Figure 8. Therefore, we may conclude that the apparent activation energy for the synthesis on thin layers of iron catalysts \vas about 28 kcal. per mole. T h e use of Equation 4 was not very effective in decreasing the spread of points in the Arrhenius plot, which suggests that catalyst activity changed moderately from point to point, or that the empirical rate equation is not valid. As the current experiments bvere made at bath temperatures from 271 ' to 294' C. over a period of several weeks? moderate changes in activity may be expected to have occurred from point to point (6).In previous tests of iron catalyst D3001 in packed beds, the activation energies increased from 19 kcal. per mole for 4to 6-mesh particles to 27.6 kcal. per mole for 28-to 32-mesh particles, and the activation energy of the surface process, without complications by diffusional resistance, was estimated to be 28 kcal. per mole (7). T h e thickness of the layer of iron on the 31-cm. cylinder, if spread evenly, was calculated to be 0.002 cm., which is smaller than the diameter of 28-to 32-mesh particles, 0.02 cm.; therefore, the activation energy of 28 kcal. per mole for the sprayed catalyst is in accord with previous work.In kinetic studies on fixed beds of catalyst D3001 reported from this laboratory (7, 9, 70), the temperatures were lower and the amount of synthesis gas reacted per unit time-weight of catalyst was smaller. Temperatures measured within the bed were usually less than 1' C. higher than the bath. These results were confirmed using the over-all heat transfer coefficient reported in the present paper. For example, for a bed of 6to 8-mesh particles operating at an hourly space velocity of 300 and a conversion of H? + CO of 65%, the average temperature differential from particle to bath was calculated to be 0.5' C.O n this basis we may conclude that the temperature of the catalyst in previous fixed-bed studies did not exceed the bath temperature substantially.REVIOUS investigators who studied the kinetics of iron oxide P reduction observed that the rate of reduction increases with increasing flow rates ( 7 , 4 ) . and that there appears to be a "critical" flo~v rate above which the rate of reduction is constant. To eliminate this variable, these workers conducted their studies using flow rates that were believed to be above the observed "critical" value. I t is believed that no detailed study of the effect of flo\v rate on reduction kinetics has been made. Many ore-reduction processes are operated under ...

No abstract