Alternative Sodium Recovery Technology?High Hydroxide Leaching: FY10 Status Report

Abstract

SummaryBoehmite leaching tests were carried out at NaOH concentrations of 10 M and 12 M, temperatures of 85°C and 60°C, and a range of initial aluminate concentrations. These data, and data obtained during earlier 100°C tests using 1 M and 5 M NaOH, were used to establish the dependence of the boehmite dissolution rate on hydroxide concentration, temperature, and initial aluminate concentration. A semiempirical kinetic model for boehmite leaching was fitted to the data and used to calculate the NaOH additions required for leaching at hydroxide concentrations between 5 M and 12 M. The optimal NaOH concentration for boehmite leaching at 85°C was estimated, based on minimizing the amount of Na that had to be added in NaOH to produce a given boehmite conversion.It was found that the sodium is used most efficiently (i.e., the mass of sodium/mass of aluminum in the waste is minimized) at NaOH concentrations of 7 to 9 M, depending on the initial distribution of the aluminum-containing compounds. The simulations indicated that the amount of sodium could potentially be reduced to one-third of that needed for the 5-M case. Furthermore, significant reductions in leaching time (per batch) were also demonstrated to be attainable at higher NaOH concentrations, typically greater than 10 M.The semi-empirical kinetic model that minimized the prediction error was found to be The fitted values of the pre-exponential constant, 0.00618/hr/(M OH)1.471 , the activation energy, 115 kJ/mol, and the exponent of the free hydroxide concentration, 1.471, were sharply defined, statistically speaking. Relatively small percentage changes in these parameters, ± 17% for the rate constant, ± 11% for the activation energy, and ± 5% for the hydroxide exponent, caused the sum of squares of the model prediction errors to double. The exponents of the boehmite conversion and initial iv aluminate saturation terms were not defined nearly as well: the error doubled only for percentage changes of 40% to 100% in these parameters. There was some indication that the boehmite exponent was higher than 2 /3, corresponding to a boehmite particle surface whose fractal dimension was between 2 and 3, but the model fit has too little sensitivity to this exponent to make a strong statement about it.In general, the model gave good predictions of reaction rate and conversion over the range of 85°C to 100°C, 5 M NaOH to 12 M NaOH, and at a 0%, 40, or 70% initial aluminate saturation. At lower temperatures or hydroxide concentrations and at higher initial aluminate saturation, the model underpredicted conversion more often than it overpredicted.v