The boehmite ‘solubility gap’

Boehmite, rather than gibbsite, precipitation has been proposed in the literature as a potential energy-saving modification of the Bayer process for the production of alumina. Previous experimental studies have reported that true equilibrium solubilities were not attained during boehmite precipitation. Instead, a pseudo-equilibrium or an apparent ‘steady-state’ aluminate concentration of about twice the boehmite solubility was reached. In this work, the dissolution and precipitation reactions in synthetic and plant liquors using seeds of (i) pure boehmite and (ii) various boehmite/gibbsite ratios were investigated at 95 °C. Only boehmite precipitation was observed on pure boehmite seed at relatively low supersaturation (alumina (A)/total caustic (TC) ≲ 0.56). The aluminate concentrations measured as a function of time decreased continuously and did not exhibit an apparent ‘steady state’. Stable equilibrium, as established by boehmite solubility measurements, was approached very slowly but not attained even after ten weeks. At higher supersaturation (A/TC ≈ 0.67), after an initial desupersaturation, ‘steady-state’ aluminate concentrations of about twice the boehmite solubility were observed. There is convincing evidence that these ‘steady states’ correspond to metastable solubility equilibria with gibbsite, which is precipitated initially and gradually transforms into the stable phase, boehmite. Gibbsite also nucleated in the case of pure boehmite seeds. ‘Steady states’ were attained in one up to several days and remained constant for one to ten days. The length of these periods correlated with the gibbsite content of the seeds. After sufficient recrystallisation of gibbsite to boehmite, the aluminate concentrations decreased significantly and eventually approached boehmite solubility, thereby following a much slower precipitation kinetics typical for boehmite. Due to short observation times, previous workers did not detect the end of the ‘steady-state’ periods and therefore failed to identify the observed ‘steady-state’ aluminate concentrations as arising from metastable solubility equilibria with gibbsite.