As a result, simulations are often restricted to small crystal surface portions, or Kossel crystals of size below 1,000 × 1,000 × 1,000 elements. Nevertheless, kMC simulations remain time-consuming, and a current restriction of the models is their limitation to handle large-size systems. On the long term, the surface geometry evolution is found to be highly dependent to the set of probabilities derived from the rate set. Then, the time and the list of elements are updated before the next iteration. Surprisingly, rough microcrystalline calcite surface dissolves globally more slowly than the is then normalized to a value close to unity, so that an element is removed at each iteration or so. Microcrystalline aragonite has the highest rate of dissolution compared to calcite. While crystal edges and convex topographies record the highest dissolution rates, the retreat is slower for flat surfaces and in topographic lows (i.e., concave areas), suggesting surface-energy related and/or diffusion-limited reactions. Evaluation of the retreat and mapping of the reaction rates at the whole surface of the samples reveals a large spatial variability in the dissolution rates, reflecting the composition and the specific contributions of the different regions of the samples. We provide a detailed 3D characterization of the geometry evolution and dissolution rate mapping at the surface of four carbonate samples, namely a calcite spar crystal, two limestone rock fragments, and an aragonite ooid, using time-lapse X-ray micro-tomography during dissolution experiments at pH 4.0. 3Géosciences Environnement Toulouse, Observatoire Midi-Pyrénées, Université Paul Sabatier, CNRS, IRD, CNES, Université de Toulouse, Toulouse, France.2ISTerre, University Grenoble Alpes, University Savoie Mont Blanc, University Gustave Eiffel, CNRS, IRD, Grenoble, France.1Departments of Geosciences and Physics, The Njord Centre, University of Oslo, Oslo, Norway.Marthe Grønlie Guren 1 François Renard 1,2 Catherine Noiriel 3 *

Convex Converts Xrd Data Files