Diffusion timescale conundrum in contact metamorphism: Oxygen isotope and trace element profiles in dolomite

Successful application of diffusion chronometry is based on detailed knowledge of diffusion parameters obtained from laboratory experiments, which also need to be upscaled to be applied to natural samples. While generally applicable, several examples of disagreement with other dating techniques and discrepancies with geological observations highlight the importance of validating diffusion parameters and their use to geochronologic application by testing them on samples from well-characterized geologic environments. The incomplete dissolution and reprecipitation reaction of dolomite associated with infiltration along grain boundaries of a fluid of magmatic origin in a contact metamorphic environment is the starting point of this study. The reaction produces dolomite rims, which are characterized by lower & delta;18O values (about 16%o) and slightly higher Fe (-1000 ppm) and Mn (-200 ppm) concentrations. The cores, in comparison, have no Fe and Mn and a & delta;18O value of about 30%o. This compositional and isotopic gradient led to chemical re-equilibration through diffusion at the reaction interface. We acquired oxygen isotope profiles using a secondary ion mass spectrometer (SIMS), with spot sizes as small as 3 & mu;m, and iron and manganese profiles with a NanoSIMS and a beam diameter of about 240 nm. These profiles were then compared against cumulative theoretical diffusion distances calculated for a modeled time - temperature path of the contact aureole using the experimentally determined diffusion parameters for O, Fe and Mn in dolomite. The oxygen isotope profiles yielded diffusion distances of ca. 10 & mu;m, corresponding to a formation temperature of about 550 degrees C, which is concordant with phase petrology and a timing for fluid infiltration at the onset of cooling of the contact aureole (i.e., non-isothermal evolution of the dolomite marbles). The Fe and Mn profiles were very sharp, with a diffusion length of about 150 nm. Such short diffusion length would be achieved at nearsurface temperature and corresponding cooling timescale are several orders of magnitude smaller than those derived from the oxygen diffusion profiles. From the mineralogy and textures, a decoupling of the oxygen isotopes and Fe-Mn reaction interface can be ruled out, suggesting that the published diffusion parameters are not fully representative, either due to the experimental conditions or by the assumed diffusion mechanism of natural processes. This case study is not the first example in the literature where the diffusion rate of Fe and Mn in natural dolomite seems to be substantially slower than predicted by experiments. Thus, there is a need to reassess the experimental diffusivities before Fe and Mn diffusion in dolomite can be reliably used to determine duration, cooling or heating paths of metamorphic processes.