B.I. Whittington

The chemistry of CaO and Ca(OH)2 relating to the Bayer process

Abstract Lime is used extensively within the Bayer process, where it performs many useful functions. In particular, it can enhance the extraction of alumina (improving the dissolution of boehmite and diaspore or the conversion of aluminogoethite), control liquor impurities (desilication, causticisation or phosphorous control), assist with the removal of impurities from the pregnant liquor (liquor ‘polishing’) and minimise soda losses in the red mud (formation of alternate desilication products or calcium titanates). This review examines these uses of lime and, where possible, uses the underlying chemistry to explain the observed benefits resulting from lime addition. A particular emphasis is placed on those factors influencing the efficiency of lime use (eg: CaO Ca(OH) 2 surface area, reaction temperature, CaO Ca(OH) 2 addition point and the presence and concentration of ‘impurities’).

The effect of reaction conditions on the composition of desilication product (DSP) formed under simulated Bayer conditions

NaCl, Na2CO3 and Na2SO4 impurities present in Bayer liquors have a detrimental effect on the efficiency of the Bayer process. However, it is possible to remove these impurities by incorporation of these salts into Bayer-sodalite or cancrinite (SOD or CAN, ideal formulae Na8(Al6Si6O24)X2·yH2O; X−=Cl−, OH−, 1/2 CO32−, 1/2 SO42−, Al(OH)4−). The Bayer-SOD or CAN is subsequently discarded with the red mud waste. This paper examines the composition of SOD or CAN formed after pre-desilication (100°C for 15 h) then digestion (150°C/30 min, 175°C/30 min, 250°C/10 min) in sodium aluminate liquors containing added Na2SO4 (0–36 g/l) and/or Na2CO3 (6–42 g/l) and/or NaCl (0–18 g/l). The results suggest that the magnitude of anion incorporation into Bayer-SOD under pseudo-Bayer conditions follows the trend: OH−≪Al(OH)4−<Cl−≤CO32−≪SO42− and that the overall SOD anion content can be approximated by: [0.5 mol Cl−/6SiO2+0.5 mol Al(OH)4−/6SiO2+moles CO32−/6SiO2+moles SO42−/6SiO2]≈1.15 [two standard deviations (2σ)=0.25]. Incorporation of SO42− into the SOD cages can best be modelled from the liquor Na2SO4 concentration by a Freundlich-type isotherm of the form (DSP molar SO3/6ReSiO2 ratio)=c1[Na2SO4]c2, where c1 and c2 are constants. By contrast, the concentrations of Cl− or CO32− incorporated into the SOD are quantitatively dependent on the concentrations of Na2SO4, Na2CO3 and NaCl present in the initial liquor. The concentration of Cl−, CO32− or SO42− in the SOD cages was observed to vary with digestion temperature. The mechanism responsible for this variation was studied. The results were found to be qualitatively consistent with a mechanism involving substantial anion incorporation during desilication at 100°C, subsequent dissolution of some of this original DSP and reprecipitation of a DSP with a composition approximately reflecting the liquor composition and temperature during digestion. The amount of reprecipitated DSP increases with temperature—experiments suggest a treatment at 150°C/30 min results in ≈10% of the SOD dissolving then reprecipitating, while treatments at 175°C/30 min or 250°C/10 min result in ≈17% and ≈50% of the SOD reprecipitating, respectively.

Pressure acid leaching of arid-region nickel laterite ore: Part I: effect of water quality

The effect of water salinity on the reactions occurring during pressure acid leaching of an arid-region laterite ore, using hypersaline water, seawater, sub-potable water and tap water, is examined. Particular emphasis is placed on the mineralogy of the residue and its implications with regard to residue volume/mass, overall acid consumption and nickel extraction. Analysis of a pressure acid leach residue by electron microprobe indicates that the residual nickel is present in phases that contain silicon and varying concentrations of aluminium, but are deficient in sulphur. Incomplete extraction of nickel from the ore may not be attributed to any one mineral phase.

Pressure acid leaching of arid-region nickel laterite ore: Part II. Effect of ore type

A comparison of the leach chemistry and residue mineralogy has been carried out on the pressure acid leaching of nontronite, limonite and saprolite ores, using hypersaline water. Results are also compared with a typical arid-region laterite feed from Bulong, which consists of a blend of these ore types. Particular emphasis is placed on the influence of ore type on liquor analysis of iron, aluminium and magnesium, residue mineralogy and nickel extraction. Microprobe evidence is presented that incomplete nickel extraction results from the presence of unreacted minor phases present in the original ore, or from the presence of nickel in the amorphous silica, in apparent association with magnesium.

Time-resolved diffraction studies into the pressure acid leaching of nickel laterite ores: a comparison of laboratory and synchrotron X-ray experiments

This paper compares time-resolved diffraction experiments that have been performed using laboratory and synchrotron X-ray sources. The experiments investigated the mechanism and kinetics of pressure acid leaching of nickel laterite ores. The sample environment was a purpose-built capillary reaction vessel, and extensive method development was conducted in the laboratory using Mo Kα radiation prior to repeating and extending the experiments at the Daresbury Synchrotron Radiation Source, beamline MPW6.2. In general, the synchrotron results confirmed the findings from previously reported laboratory work and also confirmed the presence of a minor phase that had been ambiguous in the laboratory experiments, i.e. the formation of hematite in the pressure acid leaching of saprolite. The synchrotron measurements also extended the experimental programme to include poorly diffracting laterite components that could not be examined in the laboratory, e.g. nontronite. The results from these components supported the reaction mechanisms determined from ex-situ analyses conducted in larger scale autoclaves.

Formation of lime-containing desilication product (DSP) in the Bayer process: factors influencing the laboratory modelling of DSP formation

Abstract This paper examines the effect of certain reaction variables on the mineralogy of desilication product (DSP) formed during the laboratory modelling of the reaction of kaolin and lime in synthetic Bayer liquors. In particular, the effect of the desilication (100°C) to digestion (250°C) heating rate, kaolin source and liquor composition are discussed. At a low desilication to digestion heating rate (1.6°C/min, sample cooled once the temperature reaches 250°C) the DSP mineralogy is generally unaffected by the kaolin source used or by the form in which the reactants are added (e.g.: adding the CaO, extra gibbsite and kaolin to a sodium aluminate liquor at 90°C versus adding the CaO, gibbsite and kaolin to an aluminate-free liquor at room temperature). Hydrogarnets (HG, Ca3Al2(SiO4)n(OH)(12-4n) with a silicon incorporation “n” ≈ 0.65 and a calcium-containing cancrinite (Ca-CAN) form under these conditions and the reaction pathway appears to reflect the thermodynamics of the system. By contrast, the mineralogy of the DSP formed at the higher heating rate (30°C/min, sample cooled once the temperature reaches 250°C) depends on the specific reaction conditions employed and more care must be given to the choice of initial modelling conditions. The reactant addition procedure still influences the DSP mineralogy when the time at 250°C is extended from 0 to 10 min. The implications of these results for the modelling of DSP formation using synthetic reaction mixtures are discussed. A possible mechanism for the formation of Ca-CAN is also discussed.