THE DISTRIBUTION OF DISSOLUTION RATES ON THE CALCITE CLEAVAGE SURFACE

Published rates of calcite dissolution at pH ~ 9, collected using powders or single crystals (prepared as rotating disks), range from greater than ~ 10^-9 to ~ 10^-10.2 mol/cm²/sec at room temperature. In comparison, rates derived from atomic step movements (AFM) and surface normal retreat (vertical scanning interferometry, VSI) in areas of etch pit development on freshly cleaved surfaces are lower as a group, with a lower bound of ~10^-11.6 /cm²/sec (measured by VSI at pH 8.87, 23ºC, P_CO2 10^-3.39). Here we explore the possible origins of these differences by analyzing the spatial distribution and organization of rates on the mineral surface.

On new cleavage surfaces, smooth, low relief terraces exhibit a characteristic pattern of etch pit nucleation and development. After initial nucleation, small and widely separated etch pits coalesce and cannibalize one another after several hours of reaction, ultimately resulting in large (diameter ≥ ~10² mm), intersecting etch pits. VSI rates of surface normal retreat in these areas range from 10^-11.4 to 10^-11.6 mol/cm²/sec. In contrast, areas of high macrostep density (high relief formed during cleaving) are often densely populated by small etch pits (presumably reflecting locally high defect densities) that persist even after tens of hours of reaction. These areas dissolve at VSI rates up to almost an order of magnitude faster – up to 10^-10.7 mol/cm²/sec. Although these higher rates of surface normal retreat are still slower than the slowest rates observed in powder experiments, this pattern is consistent with the notion that grain boundaries may be an important source of the overall difference in powder (bulk) versus VSI rates. It has been suggested elsewhere that the discrepancy in the rates derived from AFM step velocities versus bulk powder measurements reflects the relationship of step density and reactivity (e.g., Dove and Platt 1996). Grain edges or boundaries, along which cleavage step density must reach a maximum, would thus be sites of accelerated dissolution. Because the ratio of edge length to surface area in a free grain varies as the reciprocal of grain diameter, this leads to the expectation of a nonlinear relationship between grain size and specific dissolution rate.