Iron Silicate Slags Research Articles

Mechanism of Selenium Loss in Copper Slag

During smelting of copper sulfide concentrate, selenium is distributed between silica-saturated iron-silicate slag and copper-iron sulfide matte. The recovery coefficients of selenium between slag and matte were determined as a function of the initial concentration of selenium at 1523 K (1250 °C) under an inert atmosphere in a vertical tubular furnace. The initial concentration of selenium was varied by the addition of metallic selenium as well as selenium dioxide to the mixture of slag and matte. Analysis of the results indicated high affinity of selenium for matte. The apparent loss of selenium with the slag was attributed to the presence of selenium-enriched matte particles entrapped in the slag, rather than dissolved SeO2. The mechanisms proposed by previous investigators were discussed and also compared with the results of the present investigation.

Recoveries of rare elements Ga, Ge, In and Sn from waste electric and electronic equipment through secondary copper smelting

The recycling and recovery of valuable metals from waste materials is one of the key issues in maintaining the sustainability of base and rare metals. Especially WEEE (Waste Electric and Electronic Equipment) can be considered as a high potential resource for a number of valuable and critical metals like gallium, germanium and indium. During the mechanical processing of WEEE, these metals are primary separated into the non-ferrous scrap fractions, including copper fraction. As a consequence, the behavior of these valuable metals and the possibility of their recycling in secondary copper smelting are of great interest. This study experimentally investigates the distribution behavior of indium, gallium, germanium and tin between metallic copper and lime-free / lime-containing alumina iron silicate slags (LCu/s[Me] = [Me]Copper/(Me)Slag), as well as between solid Al-Fe spinel and slags (Lsp/s[Me] = {Me}spinel/(Me)slag). Moreover, the copper-slag-spinel equilibrium systems are examined. The experiments were executed simulating high alumina-bearing copper scrap smelting in typical black copper smelting conditions of pO2 = 10−10–10−5 atm (1 atm = 1.01325 × 105 Pa) and T = 1300 °C. The experimental technique employed utilized a highly advanced equilibration-rapid quenching method followed by Electron Probe Micro-Analysis (EPMA).The results show that tin and indium can be efficiently recovered into the copper phase in reducing process conditions (pO2 below 10−7 atm), whereas gallium dissolved preferentially into the solid spinel phase in all conditions examined. Gallium dissolution into slag and spinel was found to occur as GaO1.5, whereas indium in spinel was determined to be as InO1.5. In addition, germanium was seen to distribute preferentially into the copper phase with LCu/s[Ge] = 2–4, although its concentrations in all phases present were relatively low. Thus, the main route for germanium can be considered to be vaporization.

Distribution of Bi Between Slags and Liquid Copper

The distribution of Bi between liquid copper and calcium ferrite slag containing 24 wt pct CaO, iron silicate slag with 25 wt pct SiO2, and calcium iron silicate slags was measured at 1573 K (1300 °C) under controlled CO-CO2 atmosphere. The experimental results showed that bismuth distribution is affected by the oxygen partial pressure, and bismuth is likely to exist in slags in the 2+ oxidation state, i.e., as BiO. The distribution ratio between calcium ferrite slag and metal was found to be close to that of iron silicate slag. The Bi distribution ratio was found to decrease with increasing SiO2 and Al2O3 content in slag. Increasing temperature was found to decrease the Bi distribution ratio between slag and metal. Using the measured equilibrium data on Bi content of the metal and slag and composition dependence of the activity of Bi in liquid copper, the activity and hence activity coefficient of BiO in the slag was calculated. The close value of activity coefficient of BiO in both slags at the same oxygen partial pressure indicates that the CaO-BiO and SiO2-BiO interactions are likely to be at the same level, or the FeOx-BiO interaction is the predominant interaction for BiO in the slag. Therefore at a constant FeOx content in the slag, the CaO-BiO and SiO2-BiO interactions doesn’t affect \( \gamma_{\text{BiO}} \) significantly.

Distribution of Precious Metals (Ag, Au, Pd, Pt, and Rh) Between Copper Matte and Iron Silicate Slag

The distributions of precious metals (Ag, Au, Pd, Pt, and Rh) between copper matte and silica-saturated iron silicate slag were determined at 1523 K to 1623 K (1250 °C to 1350 °C), in controlled CO-CO2-SO2-Ar gas mixtures. The experiments were done in silica crucibles and a fixed partial pressure of sulfur dioxide for matte grades of 55, 65, and 75 wt pct Cu. High-temperature equilibration/quenching/electron probe X-ray microanalysis technique was used to obtain compositions of the equilibrated matte and slag. The technique was applied for the first time to the distributions of precious metals in simulated flash smelting conditions. The resolution of electron probe microanalysis became critical as the detection limits were insufficient to measure reliably the precious metals concentrations (except silver) in the slag. The distribution coefficient of silver, L m/s[Ag] = [wt pctAg in matte]/(wt pctAg in slag), was found to be between 200 and 300, which agrees well with the latest studies in the literature. For other precious metals, the minimum values of distribution coefficients were determined according to the detection limits in the slag. The values obtained were for gold and platinum >250, for palladium >1000, and for rhodium >900. The distribution coefficients of palladium, although locating above distribution coefficient of the detection limit, formed a clear dependency with a good repeatability as a function of the matte grade. It increased along with matte grade and was approximately 1000 at 50 pct Cu and 2000 to 3000 at 70 pct Cu. The precious metals replace metal in the matte structure and they are present as sulfides in the copper matte.

Freeze lining formation on water cooled refractory wall

The experimental study was focused into effects of a ceramic refractory lining, covering a metallic cooling element, as well as to properties and microstructures of the freeze lining formed in such cases in copper smelting and converting slags. The modified heat transfer pattern, due to the ceramic refractory lining and its air gap(s), increases effectively the average freeze lining temperature, heating its cold face from 30 to 50°C of the direct contact close to 800–900°C, even with refractory thicknesses of a few millimetres. This general feature has direct consequences to the microstructure and thickness of the freeze lining and probably also to its growth rate. The observations about effects of high copper iron silicate slag on the chemical corrosion of direct bonded magnesia–chromia refractories suggest that copper oxides are not the most aggressive components of direct to blister smelting slags. Impregnation of the fluid direct to blister slag into the open porosity of the brick, even on the cooled furnace wall, extends quickly in the initial contact several hundred micrometres into the refractory.

Equilibrium Distribution of Precious Metals Between Slag and Copper Matte at 1250–1350 °C

Metal value recoveries in extraction are the key issue for sustainability of metals. The distributions of precious metals (Ag, Au, Pd, Pt, and Rh) between copper matte (a Cu–Fe–S–O melt) and silica-saturated iron silicate slag were determined at 1250–1350 °C, under controlled oxygen and sulfur pressures and at fixed partial pressure of sulfur dioxide, in silica saturation for target matte grades of 55, 65, and 75 wt% Cu. High-temperature equilibration/quenching was performed followed by electron probe X-ray microanalysis and laser ablation–inductively coupled plasma–mass sectrometry for measuring the major elements of the matte and slag, and trace elements of the slag, respectively. The distribution coefficient of silver at 65 % matte was found to be 150, which agrees well with the most recent studies in the literature. The other values obtained were gold 1500, palladium 3000, platinum 5000, and rhodium 7000–8000. The distribution coefficients increased along with matte grade, and for palladium it was approximately 1000 at 50 % Cu and 4000–5000 at 70 % Cu. The distribution coefficients decreased along with temperature but its impact was small. The distribution mechanism of the trace elements between iron silicate slag and copper matte appear to be dominated by properties of the matte phase.

Microstructure characterisation of freeze linings formed in a copper slag cleaning slag

The initial growth rate of freeze linings on water-cooled elements submerged in molten iron silicate slag is fast. The freeze lining microstructure forming on water cooled steel surface in a high-silica, slag cleaning furnace slag of a direct-to-blister copper smelter is mostly glassy or amorphous. It contains 5-30 ?m magnetite crystals, very small and larger copper droplets as well as small magnetite and silicate nuclei embedded in the glassy silica-rich matrix. Chemically the formed freeze linings are more silica-rich than the slag from which they were generated. Magnetite (spinel) is the primary phase of the solidifying SCF slag but it does not form a continuous network through the freeze lining. Its strength is given by the intergranular silica-rich phase which initially is glassy or microcrystalline. Due to only partial slag reduction in the SCF process, large magnetite crystals are present in the freeze lining and seem to interact physically with copper droplets.

硫化製錬における非鉄金属とNaのマット－スラグ間の分配

Distribution behaviors of non-ferrous metals of copper, lead, nickel, cobalt and sodium between matte and slag were studied using slags with a composition similar to industrial copper matte smelting of low grade secondary materials. The distribution ratios of copper, lead and nickel increased with the increasing copper concentration in matte, i.e., matte grade. In contrast, those of cobalt and sodium decreased with the increasing matte grade. Two interesting effects of sodium were found out in the present study. One is to reduce copper, lead and nickel concentration in slag; and the other is suppression of lead evaporation from matte. Both effects results in the increasing distribution ratios of copper, nickel and lead; however, sodium has no influence on the distribution ratio of cobalt. The effect of silicate degree, SD, of slags on the concentration of non-ferrous metals in slag was also found out. Distribution ratios of copper, nickel, lead and cobalt in present study, in which the SDs of slags were comparatively low (0.7 ～1.3), were 2 ～4 times higher than those in the previous study using silica-saturated iron silicate slag (SD ≃ 1.5). Variation in copper and nickel concentration in slag with matte grade was also dependent on the SD. These decreased with the increasing matte grade for the slag with low silicate degree. In conclusion, we proposed that the efficient recovery of rare metals, such as nickel and cobalt, in the copper matte smelting of low grade secondary materials is expected when the influences of sodium in secondary materials and a slag with low silicate degree are effectively utilized.

Oxidation Potentials in Matte Smelting of Copper and Nickel

The oxidation potential, given as the base-ten logarithm of the oxygen partial pressure in bars and the temperature [log pO2/T, °C], defines the state of oxidation of pyrometallurgical extraction and refining processes. This property varies from copper making, [−6/1150]; to lead/zinc smelting, [−10/1200]; to iron smelting, [−13/1600]. The current article extends the analysis to the smelting of copper and nickel/copper sulfide concentrates to produce mattes of the type Cu(Ni)FeS(O) and iron silicate slags, FeOxSiO2—with oxidation potentials of [−7.5/1250].

Freeze lining formation in continuous converting calcium ferrite slags. II

The heat transfer properties of freeze linings generated in laboratory conditions, by industrial calcium ferrite slags from a continuous copper matte flash converting furnace, have been studied in situ in the molten slag using a water cooled probe technique. The measured heat conductivity of the freeze lining formed, estimated from direct measurements in steady state conditions, was 8·0±1·5 W m−1 K−1. The obtained heat conductivity of the freeze lining is 50–100% higher than that of the iron silicate slag freeze linings. The calcium ferrite slag forms a fully crystalline freeze lining. Various ferrites and metallic copper develop the observed high heat conductivity when copper precipitated from the slag during solidification fills the intergranular cavities of the ferrite crystals tightly in forming the freeze lining layer.On a étudié les propriétés de transfert de chaleur de revêtements de gel engendrés en laboratoire par des scories industrielles de ferrite de calcium, à partir d’un four éclair en continu de cémentation de matte de cuivre, dans la scorie fondue en utilisant la technique du capteur refroidi à l’eau. La conductibilité thermique mesurée du revêtement de gel formé, estimée par des mesures directes en conditions de régime permanent, était de 8·0±1·5 W m−1 K−1. La conductibilité thermique obtenue du revêtement de gel est de 50 à 100% plus élevée que celle des revêtements de gel de scorie de silicate de fer. La scorie de ferrite de calcium forme un revêtement de gel entièrement cristallin. Les différentes ferrites et le cuivre métallique développent la conductibilité thermique élevée observée lorsque le cuivre précipité à partir de la scorie pendant la solidification remplit hermétiquement les cavités intergranulaires des cristaux de ferrite dans la couche de revêtement de gel en formation.

Influence of alumina on mineralogy and environmental properties of zinc-copper smelting slags

An iron-silicate slag, from a zinc-copper smelting process, and mixtures of this slag with 5wt%, 10wt%, and 15wt% alumina addition were re-melted, semi-rapidly solidified, and characterized using scanning electron microscopy equipped with energy dispersive spectroscopy, and X-ray diffraction. The FactSage™6.2 thermodynamic package was applied to compare the stable phases at equilibrium conditions with experimental characterization. A standard European leaching test was also carried out for all samples to investigate the changes in leaching behaviour because of the addition of alumina. Results show that the commonly reported phases for slags from copper and zinc production processes (olivine, pyroxene, and spinel) are the major constituents of the current samples. A correlation can be seen between mineralogical characteristics and leaching behaviours. The sample with 10wt% alumina addition, which contains high amounts of spinels and lower amounts of the other soluble phases, shows the lowest leachabilities for most of the elements.

Influence of alumina on physical properties of an industrial zinc-copper smelting slagPart 1 – viscosity

The rotating cylinder method was applied to measure the viscosities of an industrial iron silicate slag and mixtures of this slag with 5, 10 and 15 wt-% alumina addition, in temperature range 1100–1300°C. The measured viscosities were compared with the predicted values using two of the commercially available software products for viscosity calculations, namely Thermoslag®1·5 and FactSageTM6·2. As the models can only predict viscosities for a solid free melt, obtained values by FactSageTM6·2 were modified using the Einstein–Roscoe equation. Results show that aluminium behaves as a network former cation in this type of slag, and by increasing the alumina concentration, the melt becomes progressively polymerised. Consequently, the viscosity of the slag increases at a given temperature, which is supported by thermodynamic predictions. According to the modified FactSageTM6·2 calculations, the viscosity of the solid containing slag increases from 2·1 to 5·5 poise at the industrial operating temperature (∼1250°C).

Influence of alumina on physical properties of an industrial zinc–copper smelting slag Part 3 – melting behaviour

A combination of different experimental techniques and thermodynamic calculations has been used to investigate the melting behaviour of an industrial iron silicate slag and mixtures of this slag with 5, 10 and 15 wt-% alumina addition. Differential scanning calorimetry (DSC) and thermo-optical observation were applied to monitor the solidus temperature and softening behaviour of the samples respectively. Estimation of the liquidus temperature was made using the second derivative of activation energies for viscous flow, with respect to temperature. All experimentally detected values were compared to predictions made using the FactSageTM6·2 thermodynamic package. Results show that as the slag lies in the fayalite primary phase field, the liquidus temperature decreases due to the increased alumina concentration. In the hercynite primary crystallisation phase field, however, alumina addition to the system increases the liquidus temperature. The solidus temperature does not vary significantly due to the current changes in the total alumina content of the slag.

Minimization of Copper Losses in Copper Smelting Slag During Electric Furnace Treatment

In the quest to achieve the highest metal recovery during the smelting of copper concentrates, this study has evaluated the minimum level of soluble copper in iron-silicate slags. The experimental work was performed under slag-cleaning conditions for different levels of Fe in the matte and for a range of Fe/SiO2 ratios in the slag. All experiments were carried out under conditions where three phases were present (copper–matte–slag), which is the condition typically prevailing in many slag-cleaning electric furnaces. The %Fe in the electric furnace matte was varied between 0.5 wt.% and 11 wt.%, and two different Fe/SiO2 ratios in the slag were used (targeted values were 1.4 and 1.6). All experiments were performed at 1200°C. From thermodynamic considerations, from industrial experience, and from the results obtained in this study, the minimum soluble copper content in the electric furnace slag is expected to be near 0.55 wt.% Cu. This level does not account for a portion of the copper present as mechanically entrained matte/metal droplets. Taking this into account, the current authors believe an overall copper level in discard slag between 0.7 wt.% and 0.8 wt.% can be obtained with optimal operating conditions. For these conditions, the copper losses in the slag are roughly 75% as dissolved copper and 25% as entrained matte and copper. Such conditions include operating the electric furnace at metallic copper saturation, maintaining the %Fe in the electric furnace matte between 6 wt.% and 9 wt.%, not exceeding a slag temperature of 1250°C, and controlling the Fe/SiO2 ratio in the smelting furnace slag at ≤1.5. In addition, magnetite reduction needs to be performed efficiently during the slag-cleaning cycle so as to maintain a total magnetite content of ≤7 wt.% in the discard slag. The authors further consider that under exceptionally well-controlled conditions, a copper content in electric furnace discard slag between 0.55 wt.% and 0.7 wt.% can be obtained, by minimizing entrained matte and copper solubility in the discard slag.

Effect of Slag Composition on the Distribution Behavior of Pb between FetO-SiO2 (-CaO, Al2O3) Slag and Molten Copper

The distribution behavior of Pb between molten copper and FetO-SiO2 (-CaO, Al2O3) slags was investigated at 1473 K (1200 °C) and \( p_{{{\text{O}}_{2} }} = 10^{ - 10} \,{\text{atm}} \) in view of the reaction mechanism of Pb dissolution into the slag. Furthermore, the lead capacity of the slag was estimated from the experimental results. The distribution ratio of Pb (LPb) decreases with increasing CaO content (~6 mass pct) irrespective of Fe/SiO2 ratio (1.4 to 1.7). However, the addition of alumina into a slag with Fe/SiO2 = 1.5 linearly decreases the LPb, whereas a minimum value is observed at about 4 mass pct Al2O3 at Fe/SiO2 = 1.3. The log LPb continuously decreases with increasing Fe/SiO2 ratio, and the addition of Al2O3 (5 to 15 mass pct) into the silica-saturated iron silicate slag (Fe/SiO2 < 1.0) yields the highest Pb distribution ratio. This is mainly due to a decrease in the FeO activity even at silica saturation. The log LPb linearly decreases by increasing the log (Fe3+/Fe2+) value. The Pb distribution ratio increases and the excess free energy of PbO decreases with increasing Cu2O content in the slag. However, from the viewpoint of copper loss into the slag, the silica-saturated system containing small amounts of alumina is strongly recommended to stabilize PbO in the slag phase at a low Cu2O content. The lead capacity was defined in the current study and shows a linear correlation with the activity of FeO in a logarithmic scale, indicating that the concept of lead capacity is a good measure of absorption ability of Pb in iron silicate slags, and the activity of FeO can be a good basicity index in iron silicate slag.

Direct-to-blister smelting of copper concentrates: the slag fluxing chemistry

The fundamental feature of the flash smelting process is its capability to utilise the combustion energy of sulphidic raw materials, by conducting a carefully controlled roasting and subsequent smelting in a single furnace. Its furnace and equipment design is implemented so that the reaction enthalpy released in the suspension oxidation is fully used for melting the reactive and non-reactive particles of the feed mixture. Since the first industrial application in 1940s, flash smelting has been developed for other applications than copper matte smelting, including the oxygen enriched smelting, continuous converting to blister copper with low sulphur concentration, and direct-to-blister smelting. Along with actual evolution in the processing equipment, the basic metallurgy has been under careful evaluation and development. In spite of many favourable features of iron silicate slags, they become problematic in the direct-to-blister copper smelting environments, at oxygen partial pressures higher than in matte smelting. These conditions typically locate beyond thermodynamic stability range of the homogeneous, fully molten iron silicate slags, which are more or less fully degraded ‘internally’, precipitating their solid constituents, silica and/or magnetite.

Microstructure and formation kinetics of a freeze lining in an industrial copper FSF slag

The freeze lining of an industrial copper flash smelting furnace slag, its growth kinetics and microstructure have been studied using a water cooled probe technique in a rotating crucible furnace at 1350°C. The first layers of iron silicate slag solidify on the water cooled metal surface as amorphous or glassy material with a minor fraction of crystalline spinel phase precipitated. At a distance of 4–5 mm from the cold face about 50% of the structure is composed of crystalline olivine (fayalite) and spinel phases embedded in a glassy matrix. Major thickness of the freeze lining is formed within first 15 min of slag contact with a cooled metal surface. The solidified microstructures obtained were compared with equilibrium phase assemblages calculated. The equilibrium solidification in the near solidus reactions includes the formation of pyroxene and rhodonite type phases, but they were not identified in the lining microstructures.

Thermodynamics of Arsenic in FeOx-CaO-SiO2 Slags

The distribution of arsenic between calcium ferrite slag and liquid silver (wt pct As in slag/ wt pct As in liquid silver) with 22 wt pct CaO and between iron silicate slag with 24 wt pct SiO2 and calcium iron silicate slags was measured at 1573 K (1300 °C) under a controlled CO-CO2-Ar atmosphere. For the calcium ferrite slags, a broad range of oxygen partial pressure (10–11 to 0.21 atm) was covered, whereas for the silicate slags, the oxygen partial pressure was varied from 10–9 to 3.1 × 10–7 atm. The measured relations between the distribution ratio of As and the oxygen partial pressure indicates that the oxidation state of arsenic in these slags is predominantly As3+ or AsO1.5. The measured distribution ratio of arsenic between the calcium ferrite slag and the liquid silver was about an order of magnitude higher than that of the iron silicate slag. In addition, an increasing concentration of SiO2 in the calcium-ferrite-based melts resulted in decreases in the distribution of arsenic into the slag. Through the use of measured equilibrium data on the arsenic content of the metal and slag in conjunction with the composition dependent on the activity of arsenic in the metal, the activity of AsO1.5 in the slags was deduced. These activity data on AsO1.5 show a negative deviation from the ideal behavior in these slags.

Metallurgical use of glass fractions from waste electric and electronic equipment (WEEE)

Within the European Union, it is estimated that between 8 and 9 million tons of waste electric and electronic equipment (WEEE) arises annually, of which television sets and computers account for an important part. Traditionally, Cathode Ray Tubes (CRT) have been used for TVs and computer monitors, but are rapidly being replaced by flat-screen technology. Only part of the discarded CRT glass is being recycled. Primary smelters use large amounts of silica flux to form iron-silicate slag, and can, in most cases, tolerate lead input. Use of discarded CRT glass in copper smelting is a potential alternative for utilization of the glass. The mineralogical composition of a slag sampled during ordinary slag praxis has been compared with that of a mixture of slag and CRT glass when re-melted and slowly cooled. Slag (iron-silicate slag) from Boliden Mineral AB, Sweden, was used for the experiments. Slag and glass have been mixed in various proportions: pure slag, pure glass, 90% slag-10% glass and 65% slag-35% glass, and heated in an inert atmosphere up to 1400 degrees C in a Netzsch Thermal Analysis (TA) instrument. The re-melted material has been analyzed using X-ray diffraction (XRD) and scanning electron microscopy (SEM) to determine changes in mineralogical composition after mixing with glass. The results show that the main mineralogical component of the slag is fayalite; the CRT glass is amorphous. The main crystalline phases of the slag do not change with addition of glass. An amorphous phase appears when the addition of glass is increased, which gives the sample a different structure.

Nickel, lead and antimony distributions between ferrous calcium silicate slag and copper at 1300°C

Ferrous calcium silicate (FCS) slag has been proposed as a suitable slag system for continuous copper converting. However, there is little information on the important properties of FCS slag and this must be addressed before such a slag can be properly evaluated for implementation. In this work the slag/metal distribution ratios of lead, antimony and nickel between FCS slag and copper were measured at 1300°C and an oxygen partial pressure of 10–6 atmosphere. They were found to have values of 0·93, 0·54 and 0·98 respectively. These distribution ratios were compared to reported values for calcium ferrite slag and iron silicate slags, both currently used for continuous copper converting, and predicted values for FCS slag under the same conditions. Ferrous calcium silicate slag was found to be a little more than twice as good as calcium ferrite slag at absorbing lead oxide but very similar in its ability to absorb antimony and nickel oxides. However, it was almost five times poorer than iron silicate slag for absorbing lead oxide, a little poorer for nickel oxide but almost four times better for antimony oxide. The activity coefficients of NiO, PbO and SbO1·5 in FCS slag were also calculated and found to be 4·5, 1·4 and 0·6 respectively. In terms of minor element distribution behaviour only, it is concluded that FCS slag warrants closer examination as a replacement for calcium ferrite and iron silicate slags in continuous copper converting.