The dependence of labradorite dissolution and Sr isotope release rates on solution saturation state

Abstract

Labradorite dissolution kinetics and Sr release rates were measured as a function of the saturation state of weathering solutions in column reactors. During the first 750 hours, rapid, nonstoichiometric dissolution was observed. Once steady state had been reached, both the overall dissolution and Sr release became stoichiometric. Under steady state conditions that were far from being in equilibrium with the labradorite, we measured the log of the overall labradorite dissolution rate (mol mineral/m 2/s) to be −10.6 ± 0.1 while the Sr release rate was −13.2 ± 0.1 (mol Sr/m 2/s). The isotopic ratio of the output solutions did not vary with time as both the early 87Sr/ 86Sr ratios and the later, steady state ratios were all essentially the same as that of the bulk labradorite (0.704671). As the saturation state of the solution in the columns increased from −16 to −4.5 kcal/mol, the labradorite dissolution rate decreased by a factor of ∼4.5. To quantify this decrease, we determined a function that described the dependence of the labradorite dissolution rate on the solution saturation state. Using an implicit finite difference model to predict the chemical evolution of the solution passing through the column, we found that the dependence of labradorite dissolution rate on solution saturation state that best agrees with our experimental data was Rate=−k min 0.76∗ 1− exp (−1.3×10 −17)∗( |ΔG r| RT ) 14 −0.24∗ 1− exp −0.35∗ |ΔG r| RT , where k min is the far from equilibrium rate constant (mol/m 2/s), Δ G r is the Gibbs free energy of the dissolution reaction (kcal/mol), R is the gas constant (kcal/mol K) and T is the temperature (K). The rate dependence described by this equation suggests that under far from equilibrium conditions, dissolution occurs primarily by etch pit formation at defect sites. Closer to equilibrium, etch pit formation becomes less important and dissolution becomes more uniform across the crystal surface. The dependence of the Sr release rates on solution saturation could also be described by the above equation where k min was the far from equilibrium Sr release rate. Changes in the solution saturation state did not, however, affect the isotopic ratio of the Sr released during weathering. Quantifying the rates of Sr release during plagioclase weathering and the effect that the solution saturation state has on those rates has important implications in terms of the use of Sr isotopes as a proxy for chemical weathering rates and the establishment of a more rigorous relationship between variations in the marine Sr isotopic record, average global weathering rates and atmospheric CO 2 concentrations.