A. Mitchell

The thermal characteristics of the electroslag process

The heat balance of a laboratory electroslag furnace working in quasisteady state is considered in detail. Results are presented for all possible melting mode combinations of direct and alternating (60Hz) current, with live and insulated molds, in air or argon atmospheres. The material studied is AISI 4340 steel, using a slag of CaF2 + 25 wt pct A12O3. The temperature and potential fields of the slag are determined, together with heat fluxes in the furnace. Heat balances are given for each subregion of the process, showing good agreement where results are available which permit cross checks of the balance. Suggested explanations are given for the differing behavior of the various melting modes based on variations in effective slag resistivity due to electrochemical reactions, and on variations in current path. The most important factors in determining the ingot heat balance are shown to be the electrode immersion, the slag volume’s dimensions, and the depth of the cylindrical liquid metal head on the ingot.

The electrical conductivity of some liquids in the system CaF2+CaO+Al2O3

The conductivity of slags in the binary systems CaF2+Al2O3, CaF2+CaO and the ternary system CaF2+CaO+Al2O3 has been measured, using a four-lead electrode technique at a frequency of 1 kHz. The cell design used ensured that only molybdenum metal was in contact with the slag at high temperature and that the slag was wholly contained in molybdenum. No frequency dispersion could be detected at frequencies between 0.8 to 10 kHz. It is suggested that the formation of complex ions (e.g. AlO2F23− and AlOF2−) might account for the observed effects in CaF2+Al2O3, CaF2+CaO+Al2O3 liquids as has been previously suggested. an oxide ion clustering mechanism may explain the conductivities found in CaF2+CaO.

Electrode polarization in the DC electroslag melting of pure iron

It is evident from the known ionic properties of the slags used in electroslag melting, that the dc melting process must be accompanied by Faradaic reactions on the slag/ingot and slag/electrode interfaces. The present work has determined the magnitude of the overpotentials resulting from concentration polarization at these interfaces, in the case of pure iron/CaF2+Al2O3, CaF2+CaO slags using a galvanostatic pulsing technique in an electrolytic cell. The polarization overpotential existing on an electrode in an operating ESR unit has been measured by the same technique. It is found that the potentials observed on the ESR electrode agree well with the results from the electrolytic cell. The primary anodic process is postulated to be the corrosion of iron, leading to an Fe2+-saturated layer on the anode surface at sufficiently high current densities. The cathodic process is suggested to be the Faradaic reduction of Al3+ or Ca2+, to give a concentration of [Al]Fe or (Ca)slag in the cathode interface region. This observation is supported by the fact that the cathodic potentials with respect to a C/CO reference electrode are close to those predicted from the reactions: (Al2O3)+3C=3CO(g)+2Al(l) or (CaO)+C=CO(g)+Ca(g) At very high current densities both the anodic and cathodic processes may convert to arcs, leading to process instability. The chemical and thermal effects of the overpotentials are briefly discussed and compared with the present results on ESR ingots of pure iron.

Some observations on the electrical and thermal properties of the slag-skin region in the electroslag remelting process

An unsteady-state method has been used to determine the electrical resistivity and overall heat-transfer coefficient,U, of the interface region, liquid slag/slag-skin/copper wall, in an electroslag furnace. The value ofU is found to have a slight dependence on slag temperature, slag composition, and slag-skin thickness. It is postulated that the major resistance to the transfer of heat across this composite interface lies in the discontinuity between the slag-skin and the mold wall. The numerical value ofU is found to be approximately 10−2 cal · cm−2 s−1 °C−1 for the range of conditions. The electrical resistivity of the interface is found to be a sensitive function of mold wall face temperature. The application of these results to a small ESR furnace is discussed.

Electrode temperature gradients in the electroslag process

A self-consistent model for electrode temperature gradients in an ESR process has been tested against experimental results. The results indicate that when melting steels in the electrode and mold sizes studied, the electrode material spends approximately 30 sec in the temperature gradient 1000‡C to the melting point. It is suggested that this would lead to a significant nonequilibrium retention of second-phase precipitates in a melting alloy which contained these precipitates at a lower temperature in the solid.

Some electrical characteristics of a DC electroslag unit

The various current paths possible in an electroslag unit are proposed and compared with operational results measured on a dc unit. The current found to pass through the mold walls under “mold-isolated” circuitry is small (less than 0.05 × the operating current). The metallurgical consequences of the possible operating modes are outlined and demonstrated.

Electroslag remelting with all-fluoride low conductivity slags

The electrical conductivity of liquids in the composition ranges CaF2 + 0 to 12 wt pct AIF3; CaF2 + 0 to 20 wt pct LaF3; and CaF2 + 0 to 30 wt pct YF3 has been determined at 1500° and 1600°C. It is deduced from the conductivities and the form of the phase diagrams of these systems that CaF2 + 20 wt pct YF3 is the optimum all-fluoride composition for electroslag melting or welding high melting point materials. It is demonstrated that pure iron, AISI 4340, AISI 321, and Hastelloy-X may all be electroslag melted without arcing through this slag using 60 Hz power. However, the initial postulate is confirmed in that only those materials with liquidus temperatures below that of the phase precipitated on freezing the slag can be made into ingots with good surface quality. The use of this slag in electroslag welding pure iron is investigated. It is inferred from the results that the slag composition chosen could probably be used to electroslag weld thick sections of titanium.

Heat transfer and the melting process in electroslag remelting: Part I. The behavior of small electrodes

On the assumption of one dimensional axial heat flow a mathematical model is developed for describing the temperature profile in small ESR units. The model considers the regions both above and below the slag level and allowance is made for the development of a slag crust on the electrode near the top slag level. The governing equations are solved numerically and the computed results appear to be in agreement with experimental data reported on a laboratory scale ESR unit, employing 1 in. electrodes.

The solution rate of alumina in CaF2-Al2 O3 slags

The rate of solution of A12O3 in CaF2 + 30 wt pct A12O3 (at 1518° and 1509°C) and CaF2 + 20 wt pct A12O3 (at 1500°C) liquids has been determined. The operative process is diffusion-controlled, with an interdiffusion coefficient,D for the process varying between 8.5 and 8.1 x 10-5 sq cms- 1 in the CaF2 + 30 wt pct A12O3 slags, and 4.0 × 10-5 sq cms- 1 in the CaF2 + 20 wt pct A12O3 slag. Estimations of the rate at which alumina inclusions would react with these slag during the electroslag processing of steels, indicate that electrode inclusions approaching 100 μ in diam will be dissolved.