Dynamics of altered surface layer formation on dissolving silicates

The extrapolation of mineral dissolution kinetics experiments to geological timescales has frequently been challenged by the observation that mineral dissolution rates decrease with time. In the present study, we report a detailed investigation of the early stages of wollastonite dissolution kinetics, linking time-resolved measurements of wollastonite dissolution rate as a function of crystallographic orientation to the evolution of physicochemical properties (i.e., diffusivity, density, and thickness) of amorphous silica-rich layers (ASSLs) that developed on each surface. Batch dissolution experiments conducted at room temperature and at far-from-equilibrium conditions revealed that the initial (i.e., ASSL-free) dissolution rate of wollastonite (R(hkl)) based on Ca release observe the following trend: R(010)≈R(100)>R(101)>R(001). A gradual decrease of the dissolution rate of some faces by up to one order of magnitude resulted in a modification of this trend after two days: R(010)≫R(100)⩾R(101)≈R(001). In parallel, the diffusivity of ASSLs developed on each face was estimated based on the measurement of the concentration profile of a conservative tracer (methylene blue) across the ASSL using nanoSIMS. The apparent diffusion coefficients of methylene blue as a function of the crystallographic orientation (Dapp(hkl)) observe the following trend: Dapp(010)⩾Dapp(100)>Dapp(101)≫Dapp(001), and decreases as a function of time for the (1 0 0) and (1 0 1) faces. Finally, the density of ASSL was estimated based on the modeling of X-ray reflectivity patterns acquired as a function of time. The density of ASSLs developed on the (0 1 0) faces remains low and constant, whereas it increases for the ASSLs developed on the (0 0 1) faces. On the whole, our results suggest that the impact of the formation of ASSLs on the wollastonite dissolution rate is anisotropic: while some crystal faces are weakly affected by the formation of non-passivating ASSLs (e.g., the (0 1 0) face), the dissolution of other faces is hampered by passivating ASSLs within a few hours. The observed passivation is suggested to originate from the progressive densification of the ASSL, which limits the transport of reactive species from and to the dissolving wollastonite surface, as evidenced by the estimated diffusivity of the ASSLs. Because the apparent face-specific diffusivity of the ASSLs is correlated with the face-specific initial (i.e., ASSL-free) dissolution rate of wollastonite, we propose that the extent of ASSL densification (and the resulting impact on ion transport) is (at least partly) controlled by the absolute mineral dissolution rate. Overall, this study argues that the formation and microstructural evolution of ASSLs are likely candidates for mineral ageing, highlighting the need for determining the parameters controlling the spontaneous changes of ASSL diffusivity as a function of the reaction progress.