Experimental growth of åkermanite reaction rims between wollastonite and monticellite: evidence for volume diffusion control

Abstract

Growth rates of monomineralic, polycrystalline akermanite (Ca2MgSi2O7) rims produced by solid-state reactions between monticellite (CaMgSiO4) and wollastonite (CaSiO3) single crystals were determined at 0.5 GPa dry argon pressure, 1,000–1,200°C and 5 min to 60 h, using an internally heated pressure vessel. Inert Pt-markers, initially placed at the monticellite–wollastonite interface, indicate symmetrical growth into both directions. This and mass balance considerations demonstrate that rim growth is controlled by transport of MgO. At 1,200°C and run durations between 5 min and 60 h, rim growth follows a parabolic rate law with rim widths ranging from 0.4 to 16.3 μm indicating diffusion-controlled rim growth. The effective bulk diffusion coefficient $$ D_{\text{eff,MgO}}^{\text{Ak}} $$ is calculated to 10−15.8±0.1 m2 s−1. Between 1,000°C and 1,200°C, the effective bulk diffusion coefficient follows an Arrhenius law with E a = 204 ± 18 kJ/mol and D 0 = 10−8.6±1.6 m2 s−1. Akermanite grains display a palisade texture with elongation perpendicular to the reaction interface. At 1,200°C, average grain widths measured normal to elongation, increase with the square root of time and range from 0.4 to 5.4 μm leading to a successive decrease in the grain boundary area fraction, which, however, does not affect $$ D_{\text{eff,MgO}}^{\text{Ak}} $$ to a detectible extent. This implies that grain boundary diffusion only accounts for a minor fraction of the overall chemical mass transfer, and rim growth is essentially controlled by volume diffusion. This is corroborated by the agreement between our estimates of the effective MgO bulk diffusion coefficient and experimentally determined volume diffusion data for Mg and O in akermanite from the literature. There is sharp contrast to the MgO–SiO2 binary system, where grain boundary diffusion controls rim growth.