Interdiffusivities matrix of CaO-Al2O3-SiO2 melt at 1723 K to 1823 K

Ternary oxide mixtures of lime, alumina, and silica were premelted and quenched to produce glassy cylinders. A diffusion couple was selected from the mixtures of six different compositions in such a way that the average composition could be 40 wt pct CaO-20 wt pct A1₂O₃ = 40 wt pct SiO₂. Penetration curves of the components were measured with a X-ray microprobe analyzer. The interdiffusivities matrix defined with the Matano interface has been obtained from 52 successful diffusion runs at 1723 K to 1823 K as follows; 1 $$\begin{gathered} \tilde D_{10 - 10}^{30} = 8.9 \times 10^{ - 11} \exp ( - \frac{{253,700}}{{RT}})(m^2 /s) \hfill \\ \tilde D_{10 - 20}^{30} = - 2.5 \times 10^{ - 11} \exp ( - \frac{{194,300}}{{RT}})(m^2 /s) \hfill \\ \end{gathered} $$ 2 $$\begin{gathered} \tilde D_{20 - 10}^{30} = - 4.0 \times 10^{ - 11} \exp ( - \frac{{177,600}}{{RT}})(m^2 /s) \hfill \\ \tilde D_{20 - 20}^{30} = 6.12 \times 10^{ - 11} \exp ( - \frac{{318,400}}{{RT}})(m^2 /s) \hfill \\ \end{gathered} $$ where symbols, 10, 20, and 30 mean CaO, A1₂O₃, and SiO₂, respectively, and the activation energies are in Joules per mole. The diffusion composition paths obtained are discussed in relation to Cooper’s parallelogram. The composition dependency of the above interdiffusivities is estimated from the quasibinary interdiffusivities in all composition ranges of the present oxide system in liquid state.     

Lattice and grain boundary diffusion of cerium and neodymium in nickel

Diffusion of cerium and neodymium in nickel has been studied by the serial sectioning technique using radioactive tracers¹⁴¹Ce and¹⁴⁷Nd, in the temperature ranges 700° to 1100°C for volume and 500° to 875°C for grain boundary diffusion respectively. Volume diffusivities can be expressed as: $$\begin{gathered} D_{Ce/Ni} = (0.66 \pm 0.18)\exp \left( { - \frac{{60,800 \pm 810}}{{RT}}} \right)cm^2 /\sec \hfill \\ D_{Nd/Ni} = (0.44 \pm 0.13)\exp \left( { - \frac{{59,820 \pm 830}}{{RT}}} \right)cm^2 /\sec \hfill \\ \end{gathered} $$ and grain boundary diffusivities by: $$\begin{gathered} Dg_{Ce/Ni} = 0.11\exp \left( { - \frac{{29,550}}{{RT}}} \right)cm^2 /\sec \hfill \\ Dg_{Nd/Ni} = 0.07\exp \left( { - \frac{{28,580}}{{RT}}} \right)cm^2 /\sec \hfill \\ \end{gathered} $$ Results of volume diffusion have been compared with those calculated from the theories of diffusion based on size and charge difference between the solute and the solvent atoms. Whipple and Suzuoka methods have been used to evaluate the grain boundary diffusion coefficients. Both the methods give similar results.     

Interdiffusion in the carbides of the Nb-C system

Interdiffusion coefficients in Nb₂C and NbC_1−x were measured using bulk diffusion couples in the temperature range from 1400 °C to 1700 °C. Marker experiments were used to show that carbon is the only component undergoing significant diffusion in both carbides. Carbon concentrations were measured by difference using electron probe microanalysis, and interdiffusion coefficients were taken from Boltzmann-Matano analyses of the resulting concentration profiles. This analysis clearly showed that, in NbC_1−x, interdiffusion coefficient varies with carbon concentration, and is expressed by

Self-diffusion of copper in Cu−Al solid solutions

Self-diffusion coefficients of copper in Cu?Al solid solutions in the concentration interval 0 to 19 at. pct Al and in the temperature range 800° to 1040°C have been determined by the residual activity method using the isotope Cu⁶⁴. The values of the self-diffusion coefficients in the concentration interval 0 to 14.5 at. pct Al satisfy the Arrhenius relation and their temperature dependence can be expressed by the following equations $$\eqalign{ & D_{Cu}^{Cu} = \left( {0.43_{ - 0.11}^{ + 0.15} } \right) exp \left( { - {{48,500 \pm 700} \over {RT}}} \right) cm^2 /\sec \cr & D_{Cu - 2.80 at. pct Al}^{Cu} = \left( {0.46_{ - 0.16}^{ + 0.23} } \right) exp \left( { - {{48,000 \pm 900} \over {RT}}} \right) cm^2 /\sec \cr & D_{Cu - 5.50 at. pct Al}^{Cu} = \left( {0.30_{ - 0.07}^{ + 0.09} } \right) exp \left( { - {{47,000 \pm 600} \over {RT}}} \right) cm^2 /\sec \cr & D_{Cu - 8.83 at. pct Al}^{Cu} = \left( {0.46_{ - 0.09}^{ + 0.11} } \right) exp \left( { - {{47,100 \pm 500} \over {RT}}} \right) cm^2 /\sec \cr & D_{Cu - 11.7 at. pct Al}^{Cu} = \left( {0.61_{ - 0.13}^{ + 0.17} } \right) exp \left( { - {{47,200 \pm 600} \over {RT}}} \right) cm^2 /\sec \cr & D_{Cu - 14.5 at. pct Al}^{Cu} = \left( {4.2_{ - 1.5}^{ + 2.2} } \right) exp \left( { - {{51,110 \pm 1000} \over {RT}}} \right) cm^2 /\sec \cr} $$ An analysis of the results leads to the conclusion that, in the concentration interval 0 to 11.7 at. pct Al, the frequency factor and activation enthalpy concentration dependences can be described by the following equations whereD _0Cu ^Cu and ΔH _Cu ^Cu are diffusion characteristics for self-diffusion in pure copper,X _Al is the atomic percent of aluminum, andK andB are experimental constants.     

Enthalpy of mixing of liquid Ni-Zr and Cu-Ni-Zr alloys

The partial (Δ and the integral (ΔH) enthalpies of mixing of liquid Ni-Zr and Cu-Ni-Zr alloys have been determined by high-temperature isoperibolic calorimetry at 1565 ± 5 K. The heat capacity (C _p) of liquid Ni₂₆Zr₇₄ has been measured by adiabatic calorimetry (C _p=53.5±2.2 J mol⁻¹ K⁻¹ at 1261±15 K). The integral enthalpy of mixing changes with composition from a small positive (Cu-Ni, ΔH (x _Ni=0.50, T=1473 to 1750 K)=2.9 kJ mol⁻¹) to a moderate negative (Cu-Zr; ΔH(x _Zr=0.46, T=1485 K)=−16.2 kJ mol⁻¹) and a high negative value (Ni-Zr; ΔH(x _Zr=0.37, T=1565 K)=−45.8 kJ mol⁻¹). Regression analysis of new data, together with the literature data for liquid Ni-Zr alloys, results in the following relationships in kJ mol⁻¹ (standard states: Cu (1), Ni (1), and Zr (1)):for Ni-Zr (1281≤T≤2270 K),

Alumina solubility in molten salt systems of interest for aluminum electrolysis and related phase diagram data

The solubility of alumina in molten Na₃AlF₆ containing various amounts of AlF₃, CaF₂, and LiF was determined by measuring the weight loss of a rotating sintered corundum disc. The results were fitted to the following empirical expression: 1 $$ [Al_2 O_3 ]_{sat} = A\left( {\frac{t} {{1000}}} \right)^B $$ where 2 $$ \begin{gathered} A = 11.9 - 0.062[AlF_3 ] - 0.003[AlF_3 ]^2 - 0.50[LiF] \hfill \\ - 0.20[CaF_2 ] - 0.30[MgF_2 ] + \frac{{42[LiF] \cdot [AlF_3 ]}} {{2000 + [LiF] \cdot [AlF_3 ]}} \hfill \\ B = 4.8 - 0.048[AlF_3 ] + \frac{{2.2[LiF]^{1.5} }} {{10 + [LiF] + 0.001[AlF_3 ]^3 }} \hfill \\ \end{gathered} $$ where the square brackets denote weight percent of components in the system Na₃AlF₆-Al₂O₃ (sat)-AlF₃-CaF₂-MgF₂-LiF and t is the temperature in degree Celsius. The standard deviation between the equation and the experimental points in the temperature range from 1050 °C to about 850 °C was found to be 0.29 wt pct Al₂O₃. A series of revised phase diagram data of interest for aluminum electrolysis was derived based on the present work and recently published data for primary crystallization of Na₃AlF₆ in the same systems.     

Iron redox equilibria in CaO-Al2O3-SiO2 and MgO-CaO-Al2O3-SiO2 slags

Measurements have been made of the ratio of ferric to ferrous iron in CaO-Al₂O₃-SiO₂ and MgO-CaO-Al₂O₃-SiO₂ slags at oxygen activities ranging from equilibrium with pCO₂/pCO≈0.01 to as high as air at temperatures of 1573 to 1773 K. At 1773 K, values are given by $\begin{gathered} \log {\text{ }}\left( {\frac{{Fe^{3 + } }}{{Fe^{2 + } }}} \right) = 0.3( \pm {\text{ }}0.02){\text{ }}Y + {\text{ }}0.45( \pm {\text{ }}0.01){\text{ }}\log \hfill \\ \left( {\frac{{pCO_2 }}{{pCO}}} \right) - 1.24( \pm {\text{ }}0.01) \hfill \\ \end{gathered} $ where Y=(CaO+MgO)/SiO₂, for melts with the molar ratio of CaO/SiO₂=0.45 to 1.52, 10 to 15 mol pct Al₂O₃, up to 12 mol pct MgO (at CaO/SiO₂≈1.5), and with 3 to 10 wt pct total Fe. Available evidence suggests that, to a good approximation, these redox equilibria are independent of temperature when expressed with respect to pCO₂/pCO, probably from about 1573 to 1873 K. Limited studies have also been carried out on melts containing about 40 mol pct Al₂O₃, up to 12 mol pct MgO (at CaO/SiO₂≈1.5), and 3.6 to 4.7 wt pct Fe. These show a strongly nonideal behavior for the iron redox equilibrium, with $\frac{{Fe^{3 + } }}{{Fe^{2 + } }} \propto \left( {\frac{{pCO_2 }}{{pCO}}} \right)^{0.37} $ The nonideal behavior and the effects of basicity and Al₂O₃ concentration on the redox equilibria are discussed in terms of the charge balance model of alumino-silicates and the published structural information from Mössbauer and NMR (Nuclear Magnetic Resonance) spectroscopy of quenched melts.     

Determination of standard gibbs energies of formation of Cr2N and CrN

The standard Gibbs energies of formation of Cr₂N and CrN have been measured by an equilibration technique and by using thermogravimetry and differential thermal analysis (TG-DTA) at temperatures ranging from 1232 to 1523 K. The results are expressed as follows:

Mass spectrometric determination of activities for alloys with complex vapor species: Bi-Pb and Bi-Tl

For solutions from which complex species vaporize (Bi₂, Si₂, Al₂O, Sb₄, and so forth) new methods of determining the thermodynamic properties from mass spectrometric data are demonstrated. In order to test the feasibility of these new techniques, experiments have been carried out on the liquid Bi-Pb and Bi-Tl systems for which adequate thermodynamic data are available. In evaluating the thermodynamic properties, the ion current ratiosI _Pb ⁺/IBi₂/+ andI _Tl ⁺/IBi₂/+ were employed,e.g. $$\log {\text{ }}\gamma _{{\text{Bi}}} {\text{ = - }}\mathop {\int {\frac{{N_{Pb} }}{{1{\text{ + }}N_{Pb} }}d} }\limits_{N_{Bi} = 1}^{N_{{\text{Bi}}} = N_{Bi} } {\text{ }}\left\{ {{\text{log}}\frac{{{\text{1}}_{{\text{Pb}}}^{\text{ + }} {\text{ }}N_{Bi}^2 }}{{I_{Bi2}^ + {\text{ }}N_{Pb} }}} \right\}$$ Measuring these particular ion current ratios eliminates errors resulting from the fragmentation of the complex vapor species in evaluating the thermodynamic properties. A dimer-monomer technique, which corrects for fragmentation, was also demonstrated. The results using these two independent approaches are in good agreement with each other as well as with previous investigations. The activity coefficients in both systems adhere to the quadratic formalism over large composition ranges,e.g. $$\begin{gathered} \log {\text{ }}\gamma _{{\text{Pb}}} {\text{ = - 0}}{\text{.255 }}N_{Bi}^2 {\text{ }}N_{{\text{Bi}}} {\text{< 0}}{\text{.8}} \hfill \\ \log {\text{ }}\gamma _{{\text{Tl}}} {\text{ = - 0}}{\text{.805 }}N_{Bi}^2 {\text{ }}N_{{\text{Bi}}} {\text{< 0}}{\text{.7}} \hfill \\ \end{gathered} $$      

The Solubility and Diffusivity of Fluorine in Solid Copper from Electrochemical Measurements

The solubility and diffusivity of fluorine in solid copper were determined electrochemically using the double solid-state cell $$Ni + NiF_2 \left| {CaF_2 } \right|Cu\left| {CaF_2 } \right|Ni + NiF_2 .$$ In the temperature range 757 to 920°C, the diffusivity of fluorine in solid copper was found to be $$D_F \left( {{{cm^2 } \mathord{\left/ {\vphantom {{cm^2 } s}} \right. \kern-\nulldelimiterspace} s}} \right) = 9.32 \times 10^{ - 2} \exp \left( {\frac{{ - 98,910 {J \mathord{\left/ {\vphantom {J {mole}}} \right. \kern-\nulldelimiterspace} {mole}}}} {{RT}}} \right).$$ . The results obtained for the dissolution of fluorine as atoms in solid copper showed large scatter. However, the equilibrium dissolution of fluorine follows Sieverts’ law. Above the melting point (770°C) of CuF₂, the mean solubility of fluorine in solid copper, for the equilibrium Cu(s)+ CuF ₂(l), follows the relationship $$N_F^s (atom fraction) = 0.98 \exp \left( {\frac{{ - 79,500 {J \mathord{\left/ {\vphantom {J {mole}}} \right. \kern-\nulldelimiterspace} {mole}}}} {{RT}}} \right).$$      

The oxygen potential of the systems Fe+FeCr2O4+Cr2O3 and Fe+FeV2O4+V2O3 in the temperature range 750–1600°C

From electromotive force (emf) measurements using solid oxide galvanic cells incorporating ZrO₂-CaO and ThO₂?YO_1.5 electrolytes, the chemical potentials of oxygen over the systems Fe+FeCr₂O₄+Cr₂O₃ and Fe+FeV₂O₄+V₂O₃ were calculated. The values may be represented by the equations: $$\begin{gathered} 2Fe\left( {s,1} \right) + O_2 \left( g \right) + 2Cr_2 O_3 \left( s \right) \to 2FeCr_2 O_4 \left( s \right) \hfill \\ \Delta \mu _{O_2 } = - 151,400 + 34.7T\left( { \pm 300} \right) cal \hfill \\ = - 633,400 + 145.5T\left( { \pm 1250} \right) J \left( {750 to 1536^\circ C} \right) \hfill \\ \Delta \mu _{O_2 } = - 158,000 + 38.4T\left( { \pm 300} \right) cal \hfill \\ = - 661,000 + 160.5T\left( { \pm 1250} \right) J \left( {1536 to 1700^\circ C} \right) \hfill \\ 2Fe\left( {s,1} \right) + O_2 \left( g \right) + 2V_2 O_3 \left( s \right) \to 2FeV_2 O_4 \left( s \right) \hfill \\ \Delta \mu _{O_2 } = - 138,000 + 29.8T\left( { \pm 300} \right) cal \hfill \\ = - 577,500 + 124.7T\left( { \pm 1250} \right) J \left( {750 to 1536^\circ C} \right) \hfill \\ \Delta \mu _{O_2 } = - 144,600 + 33.45T\left( { \pm 300} \right) cal \hfill \\ = - 605,100 + 140.0T\left( { \pm 1250} \right) J \left( {1536 to 1700^\circ C} \right) \hfill \\ \end{gathered} $$ . At the oxygen potentials corresponding to Fe+FeCr₂O₄+Cr₂O₃ equilibria, the electronic contribution to the conductivity of ZrO₂?CaO electrolyte was found to affect the measured emf. Application of a small 60 cycle A.C. voltage with an amplitude of 50 mv across the cell terminals reduced the time required to attain equilibrium at temperatures between 750 to 950°C by approximately a factor of two. The second law entropy of iron chromite obtained in this study is in good agreement with that calculated from thermal data. The entropies of formation of these spinel phases from the component oxides can be correlated to cation distribution and crystal field theory.     

The thermodynamic properties of liquid Fe−Si alloys

The thermodynamic properties of liquid Fe?Si alloys have been determined electrochemically by use of the following galvanic cells: $$\begin{gathered} Cr - Cr_2 O_3 (s)|ZrO_2 (CaO)|Fe - Si(l), SiO_2 (s) \hfill \\ Cr - Cr_2 O_3 (s)|ThO_2 (Y_2 O_3 )|Fe - Si(l), SiO_2 (s) \hfill \\ \end{gathered} $$ The free energy of formation of SiO₂ was measured and is ?139.0 and ?134.3 kcals per mole at 1500° and 1600°C, respectively. The activity coefficients of iron and silicon for the atom fraction of siliconN _Si<0.35 at 1600° and 1500°C can be represented by the quadratic formalism. $$\begin{gathered} \left. {\begin{array}{*{20}c} {log \gamma _{Fe} = - 2.12 N_{Si}^2 } \\ {log \gamma _{Si} = - 2.12 N_{Fe}^2 - 0.22} \\ \end{array} } \right\}1600^ \circ C (2912^ \circ F) \hfill \\ \left. {\begin{array}{*{20}c} {log \gamma _{Fe} = - 2.50 N_{Si}^2 } \\ {log \gamma _{Si} = - 2.50 N_{Fe}^2 - 0.13} \\ \end{array} } \right\}1500^ \circ C (2732^ \circ F) \hfill \\ \end{gathered} $$ The results indicate that an excess stability peak occurs at about the equimolar composition. Combining the heats of solution determined in this study with previous data indicates that the heats also follow the quadratic formalism. The partial molar heats, \(\bar L_{Si} \) and \(\bar L_{Fe} \) , are represented by $$\begin{gathered} \bar L_{Si} = - 31 N_{Fe}^2 - 4 kcals per mole \hfill \\ \bar L_{Fe} = - 31 N_{Si}^2 kcals per mole \hfill \\ \end{gathered} $$ ForN _Si less than 0.35 and by $$\begin{gathered} \bar L_{Si} = - 22 N_{Fe}^2 \hfill \\ \bar L_{Fe} = - 22 N_{Fe}^2 - 7.0 \hfill \\ \end{gathered} $$ forN _Fe less than 0.35. There is an inflection point in the transition region similar to an excess stability peak for the excess free energies. At 1600°C the ThO₂(Y₂O₃) electrolyte exhibited insignificant electronic conductivity at oxygen partial pressures as low as that in equilibrium with Si?SiO₂ (2×10^?16 atm).     

Kinetics of gold(III) chloride complex reduction using sulfur(IV)

In this article, the effect of different kinetic parameters such as pH, temperature, gold, and reductant concentrations on the rate of Au reduction from aqueous chloride solutions by NaHSO₃ is investigated. On the basis of available experimental data, the possible mechanism of [AuCl₄]⁻ reduction by sulfur(IV) is also assumed. The suggested mechanism yields the rate equation for reduction of [AuCl₄]⁻, which is given in the form

Thermodynamic behavior in binary metallic solutions

The applicability of Krupkowski’s formalism $$\begin{gathered} ln \gamma _1 = \omega \left( T \right)\left( {1 - X_1 } \right)^m \hfill \\ ln \gamma _2 = \omega \left( T \right)\left[ {\left( {1 - X_1 } \right)^m - \frac{m}{{m - 1}}\left( {1 - X_1 } \right)^{m - 1} + \frac{1}{{m - 1}}} \right] \hfill \\ \end{gathered} $$ in interpreting experimental data is shown for several binary systems. Both dilute and concentrated solutions are considered. In dilute solutions (Henry’s law region) these equations exclude constant values of the activity coefficients. These formulae withm>1 satisfy Raoults law and Henry’s law as limiting cases. However, experimental data indicate that only in two systems, namely Zn-Sn and Zn-Bi,γ _Zn ⁰ =γ _Zn over a finite composition range. Whenm is close to unity, as is the case for the Zn-Sn and Zn-Bi systems Raoult’s law is not satisfied untilX _Zn is infinitesimally close to unity. Data for concentrated zinc solutions for both systems support this conclusion. A comparison of Krupkowski’s method with Darken’s quadratic formalism was also carried out, and it was shown that both methods give similar results whenm=2.     

The Diffusion Coefficient of Scandium in Dilute Aluminum-Scandium Alloys

The diffusion coefficient of Sc in dilute Al-Sc alloys has been determined at 748 K, 823 K, and 898 K (475 °C, 550 °C, and 625 °C, respectively) using semi-infinite diffusion couples. Good agreement was found between the results of the present study and both the higher temperature, direct measurements and lower temperature, indirect measurements of these coefficients reported previously in the literature. The temperature-dependent diffusion coefficient equation derived from the data obtained in the present investigation was found to be \( D \left( {{\text{m}}^{2} /{\text{s}}} \right) = \left( {2.34 \pm 2.16} \right) \times 10^{ - 4} \left( {{\text{m}}^{2} /{\text{s}}} \right) { \exp }\left( {\frac{{ - \left( {167 \pm 6} \right) \left( {{\text{kJ}}/{\text{mol}}} \right)}}{RT}} \right). \) Combining these results with data from the literature and fitting all data simultaneously to an Arrhenius relationship yielded the expression \( D \left( {{\text{m}}^{2} /{\text{s}}} \right) = \left( {2.65 \pm 0.84} \right) \times 10^{ - 4} \left( {{\text{m}}^{2} /{\text{s}}} \right) { \exp }\left( {\frac{{ - \left( {168 \pm 2} \right) \left( {{\text{kJ}}/{\text{mol}}} \right)}}{RT}} \right). \) In each equation given above, R is 0.0083144 kJ/mol K, T is in Kelvin, and the uncertainties are ±1 standard error.     

Equilibrium of calcium vapor with liquid iron and the interaction of third elements

The dissolution equilibrium of calcium vapor in liquid iron was carried out at 1873 K in a two-temperature zone furnace using a vapor-liquid equilibration method. A sealed Mo reaction chamber and a self-made CaO crucible were used in this study. The thermodynamic parameters obtained are as follows. For reaction Ca (g)=[Ca],

Determination of Gibbs Energy of Formation of Molybdenum-Boron Binary System by Electromotive Force Measurement Using Solid Electrolyte

The standard Gibbs energies of formation of Mo₂B, ??MoB, Mo₂B₅, and MoB₄ in the molybdenum-boron binary system were determined by measuring electromotive forces of galvanic cells using an Y₂O₃-stabilized ZrO₂ solid oxide electrolyte. The results are as follows: $$ \begin{aligned} \Updelta_{\text{f}} {\text{G}}^\circ \left( {{\text{Mo}}_{2} {\text{B}}} \right)/{\text{J}}\,{\text{mol}}^{ - 1} & = - 193100 + 44.10T \pm 700\left( {1198{\text{ K to }}1323{\text{ K}}\left( {925^\circ {\text{C to }}1050^\circ {\text{C}}} \right)} \right) \\ \Updelta_{\text{f}} {\text{G}}^\circ (\alpha {\text{MoB}})/{\text{J}}\,{\text{mol}}^{ - 1} & = - 164000 + 26.45T \pm 700\left( {1213{\text{ K to }}1328{\text{ K}}\left( {940^\circ {\text{C to }}1055^\circ {\text{C}}} \right)} \right) \\ \Updelta_{\text{f}} {\text{G}}^\circ \left( {{\text{Mo}}_{2} {\text{B}}_{5} } \right)/{\text{J}}\,{\text{mol}}^{ - 1} & = - 622500 + 117.0T \pm 3000\left( {1205{\text{ K to }}1294{\text{ K}}\left( {932^\circ {\text{C to }}1021^\circ {\text{C}}} \right)} \right) \\ \Updelta_{\text{f}} {\text{G}}^\circ \left( {{\text{MoB}}_{4} } \right)/{\text{J}}\,{\text{mol}}^{ - 1} & = - 387300 + 93.53T \pm 3000\left( {959{\text{ K to }}1153{\text{ K}}\left( {686^\circ {\text{C to }}880^\circ {\text{C}}} \right)} \right) \\ \end{aligned} $$ where the standard pressure is 1 bar (100 kPa).     

The kinetics of formation of h2o and co2 during iron oxide reduction

The kinetics of the chemical reaction-controlled reduction of iron oxides by H₂/H₂O and CO/CO₂ gas mixtures are discussed. From an analysis of the systems it is concluded that the decomposition of the oxides takes place by the two dimensional nucleation and lateral growth of oxygen vacancy clusters at the gas/oxide interface. The rates of decomposition of the oxides under conditions of chemical reaction control are dependent not only on the partial pressures of the reacting gases at the reaction temperature but also on the oxygen activity of the prevailing atmosphere. Application of this model to the kinetic data leads to the determination of the maximum chemical reaction rate constants for the decomposition of the iron oxide surfaces. Assuming the reactions H₂ (g) + O(ads) → H₂O(g) andCO(g) + O(ads) → CO₂ (g) to be rate controlling the maximum chemical reaction rate constants for the reduction of iron oxides are given by $$\Phi _{{\text{H}}_{\text{2}} } = 10^{.00} exp \left( {\frac{{ - 69,300}}{{RT}}} \right)mol m^{ - 2} s^{ - 1} atm^{ - 1} $$ and $$\Phi _{CO} = 10^{4.40} \exp \left( {\frac{{103,900}}{{RT}}} \right)mol m^{ - 2} s^{ - 1} atm^{ - 1} $$ The maximum chemical reaction rate constants do not necessarily indicate the maximum rates which can be achieved in practice since these will depend on the limitations imposed by mass transport in the systems. The rate constants are important however since they indicate for the first time the upper limit of any reduction rate in these systems. The fractions of reaction sites which appear to be active on wüstite surfaces in equilibrium with iron are calculated. A direct relationship between chemical reaction rates on liquid iron surfaces and rates on atomically rough iron oxide surfaces is postulated.