We have measured the surface controlled dissolution rates of natural calcium carbonate minerals (limestone and marble) in H2O-CO2 solutions by using free drift batch experiments under closed system conditions with respect to CO2, at 10 degrees C with an initial partial pressure of carbon dioxide of 5.10(-2) atm. All experiments revealed reaction rates F, which can be described by the empirical relation: F-n1 = k(n1) . (1 - c/c(eq))(n1) for c < c(s), which switches to a higher order n(2) for calcium concentrations c greater than or equal to c(s) described by F-n2 = k(n2) . (1 - c/c(eq))(n2) . k(n1) and k(n2) are rate constants in mmole/(cm(2) . s), c(eq) is the equilibrium concentration with respect to calcite. The values of the constants n(1), n(2), k(n1), k(n2) and c(s) depend on the V/A ratio employed, where V is the volume of the solution and A is the surface area of the reacting mineral. Different calcium carbonate minerals exhibit different values of the kinetic constants. But generally with increasing V/A, there is a steep variation in the values of all kinetic constants, such that the rates are reduced with increasing V/A ratio. Finally with sufficiently large V/A these values become constant. These results are explained by assuming intrinsic inhibitors in the bulk of the mineral. During dissolution these are released from the calcite matrix and are adsorbed irreversibly at the reacting surface, where they act as inhibitors. The thickness d of the mineral layer removed by dissolution is proportional to the VIA ratio. The amount of inhibitors released per surface area is given by d c(int), where c(int) is their concentration id the bulk of the mineral. At low thicknesses up to approximate to 3 . 10(-4) cm in the investigated materials, the surface concentration of inhibitors increases until saturation is attained for thicknesses above this value. To analyze the surface concentration and the type of the inhibitors we have used Auger spectroscopy, which revealed the presence of aluminosilicate complexes at the surface of limestone, when a thickness of d approximate to 10(-3) cm had been removed by dissolution. In unreacted samples similar signals, weaker by one order of magnitude, were observed. Depth profiles of the reacted sample obtained by Ar-ion sputtering showed the concentration of these complexes to decrease to the concentration observed in the unreacted sample within a depth of about 10 nm. No change of the concentration with depth was observed in unreacted samples. These data suggest that complexes of aluminosilicates act as inhibitors, although other impurities cannot be excluded. Copyright (C) 1999 Elsevier Science Ltd.
