A literature survey of the experimentally determined values for the surface tension γ of molten pure metals has been carried out. It is intended to provide an assessed data base for use with a mathematical model currently being developed at the National Physical Laboratory, UK, to predict the surface tension of molten metal alloys.

Review of data for the surface tension of pure metals

The equilibrium interfacial energy between a liquid iron alloy and a liquid slag is a key physical parameter in the design of steel-refining processes as high interfacial energies are desired to avoid emulsification of slag in steel and the creation of casting defects. During a chemical reaction between a liquid iron alloy droplet and a liquid slag, it is possible to observe by X-ray photography a number of dynamic interfacial phenomena such as droplet flattening, interfacial turbulence, and spontaneous emulsification that can potentially lead to serious processing problems. These dynamic phenomena have been studied during reactions between Fe-Al and Fe-Ti alloys and silica-containing slags, and the presence of significant interfacial disturbance has been observed during the times of high reaction rate between the slag and the metal. It is suggested that interfacial chemical reactions induce Marangoni and natural convection at the slag-metal interface. This interfacial flow gives rise to interfacial waves due to a Kelvin-Helmholtz instability. The waves grow, become unstable, and lead to spontaneous emulsification of slag in steel and steel in slag. Experiments using industrial samples and controlled laboratory tests have indicated that this phenomenon may be more common than once thought and could lead to some serious problems in the processing of steel alloys containing high quantities of aluminum and/or titanium.

Dynamic and equilibrium interfacial phenomena in liquid steel-slag systems

Together with previous values at 1873 K, new measurements of the enthalpies of formation of liquid Cu-Fe alloys, carried out by high-temperature calorimetry at 1673 K, have revealed a temperature dependence of ΔH, corresponding to a negative excess heat capacity of mixing Our calorimetric results were combined with critically selected values of the enthalpies of alloying and activities of components in liquid and solid solutions, as well as parameters of stable and metastable phase transitions, to perform a thermodynamic evaluation of the system The latter was carried out in the spirit of the CALPHAD approach, using Thermo-Calc software The thermodynamic model generates a self-consistent description of all thermodynamic properties and phase equilibria, including metastable solidification, in close agreement with reliable experimental data In particular, a very satisfactory representation of stable (L+δ)/γ, (L + γ)/e, γ/(e + α) and metastable (L1/L2), (L1 + L2)/e phase boundaries has been achieved The occurrence of a monotectic reaction L2 ↔ L1 + δ is suggested in the region of the metastable precipitation of δ-phase A comprehensive thermodynamic model of the Cu-Fe system is used to explain the widening of the concentration limits of formation of supersaturated bcc and fcc solutions prepared by highly nonequilibrium methods of synthesis

Thermodynamics of alloys and phase equilibria in the copper-iron system

Oxidation state of iron in silicate glasses and melts: a thermochemical model

https://openresearch-repository.anu.edu.au/bitstream/1885/139076/1/1-s2.0-S0009254117306782-main.pdf

Zinc isotope composition of the Earth and its behaviour during planetary accretion

Measurements of the rates of dissociation of CO2 on liquid iron oxides and CaO-saturated liquid calcium ferrites have been made by the14CO2-CO isotope exchange technique. For temperatures up to about 1550 °C, the apparent first order rate constants for both melts are essentially inversely proportional to the equilibrium CO2/CO ratio over the range studied (≈0.4 to 12). Evidence is presented that the rate limiting step in the interfacial oxidation of these melts by CO2 is the dissociation of CO2- Rates of oxidation, in mol cm−2 s−1, in CO2-CO atmospheres are deduced to be given by the equations:v =(
 p
 CO
 2
 a
 −
 p
 CO) exp(−15,900/T − 2.03) andv = (pCO2
 α
 o
 -1
 −pCO) exp(−3800/T − 6.93) for the iron oxides and CaO-saturated calcium ferrites, respectively, wherea
 0
 is the oxygen activity of the melt expressed as the equilibrium CO2/CO ratio and the pressures are in atmospheres. The strong dependence on the CO2/CO ratio is shown to be consistent with the need to transfer two charges to the adsorbing or dissociating CO2 molecule. Correlation with the existing surface tension data is inconclusive.

Isotope exchange studies of the rate of dissociation of CO2 on liquid iron oxides and CaO-saturated calcium ferrites

Measurements of the rates of dissociation of CO2 have been made by the14CO2-CO isotope exchange technique on liquid iron silicates, calcium iron silicates, and silica-saturated manganese silicates as functions of temperature and imposed equilibrium CO2/CO ratio. It is shown that the rates of reduction of the liquid iron silicates and an iron oxide-rich slag in CO-CO2 atmospheres are consistent with the rates of isotope exchange, indicating a common rate determining step. The dependences of the apparent first order rate constant on the oxygen activity for the dissociation of CO2 on silica-saturated iron silicates and an equimolar “FeO”-CaO-SiO2 melt are found to be closely consistent with the ability to transfer two charges to the adsorbing or dissociating CO2 molecule, as was previously found for liquid iron oxide and lime-saturated calcium ferrites. Apparent rate constants at fixed oxygen activity are found to increase generally with the basicities of the melts.

Interfacial rates of reaction of CO2 with liquid iron silicates, silica-saturated manganese silicates, and some calcium iron silicates

The surface tensions of melts in the system CaO-FeO-SiO2 have been measured in the temperature range 1573 to 1708 K using the hollow cylinder technique. The iron oxide content was maintained constant at 30 wt pct and the CaO/SiO2 wt pct ratio was varied in the range 0.43 to 1.5. Surface tension increases with increasing basicity and with decreasing temperature. The data were used to test published correlations of slag foaming indexes with surface tension and viscosity. Foam life increases with increasing viscosity and with decreasing surface tension.

The surface tensions and foaming behavior of melts in the system CaO-FeO-SiO2

The activity of ZnO in ZnO-SiO2 and CaO-ZnO-SiO2 melts has been measured at 1560 °C using a transpiration technique with CO-CO2 mixtures as the carrier gas. The activities of ZnO in dilute solution in 42 wt pct SiO2−38 wt pct CaO-20 wt pct A12O3 in the range 1400° to 1550 °C and in 62 wt pct SiO2−23.3 wt pct CaO−14.7 wt pct A12O3 at 1550 °C have also been measured. The measured free energies of formation of ZnO-SiO2 melts are significantly more negative than published estimated values and this, together with the behavior observed in the system CaO-Al2O3-SiO2, indicate that ZnO is a relatively basic oxide. The results are discussed in terms of the polymerization model of binary silicate melts and ideal silicate mixing in ternary silicate melts. The behavior of ZnO in dilute solution in CaO-Al2O3-SiO2 melts is discussed in terms of the possibility of the fluxing of ZnO by iron blast furnace slags.

The thermodynamic activity of ZnO in silicate melts

Padday’s cone technique allows determination of the surface tension of a liquid from measurement of the maximum excess force exerted on a cone during its immersion in, or withdrawal from, the liquid and knowledge of the density of the liquid. As the technique does not require rupture of the meniscus, an equilibrium measurement is obtained. The surface tensions of Fe-saturated iron silicate melts and Fe-saturated calcium ferrites, measured at 1410 °C using iron cones, are in good agreement with values in the literature obtained using the hollow cylinder technique. The maximum excess forces exerted on Pt-Rh cones immersed in liquid iron oxides at 1460 °C in the composition range from saturation with iron toXFe2O3 = 0.205 have been measured. The surface tensions of these melts cannot be determined unambiguously from the experimental measurements because of an uncertainty in the densities of the melts. However, it is shown that the literature values for the surface tension of liquid iron oxide are incorrect because of an error in the experimental procedures used.

D. R. Gaskell

Papers

Isotope exchange studies of the rate of dissociation of CO2 on liquid iron oxides and CaO-saturated calcium ferrites

Interfacial rates of reaction of CO2 with liquid iron silicates, silica-saturated manganese silicates, and some calcium iron silicates

The surface tensions and foaming behavior of melts in the system CaO-FeO-SiO2

The thermodynamic activity of ZnO in silicate melts

Measurement of the surface tensions of Fe-saturated iron silicate and Fe-saturated calcium ferrite melts by Padday’s cone technique