Project Details
Projekt Print View

Diffusion of H bearing species in silicate glasses at low temperatures - development of a new experimental technique

Subject Area Mineralogy, Petrology and Geochemistry
Term from 2017 to 2024
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 342185776
 
Final Report Year 2022

Final Report Abstract

We developed a new method to determine hydrogen diffusion rates in melts, glasses, and minerals at low temperatures. A combination of thin film diffusion couples (for silicate glasses) or ion implantation (for clinopyroxene (cpx)) with high resolution H depth profiling by Nuclear Resonance Reaction Analysis (NRRA) made it possible to study diffusion profiles of H in these materials from experiments in different devices (e.g., Cold Seal Pressure Vessel (CSPV) for glasses and gas-mixing furnace for cpx). The richness of information we gained from the different experiments provides details on how diffusion is affected by different boundary conditions. Values for diffusion coefficients (150 °C - 600 °C for glasses and 195 °C - 400 °C for cpx) were obtained by numerical modelling (finite difference) of measured diffusion profiles. This method allowed to consider complex boundary conditions in the model. We used our results to test already existing diffusion models that involve different concentration dependencies and speciation reactions in silicate glasses and melts. A new diffusion-speciation-reaction model was introduced for hydrogen in silicate glasses based on our results. In contrast to previous models, it assigns OH (commonly believed to be immobile) a diffusion coefficient that represents an observable contribution of OH to the total net flux of H, at temperatures > 450 °C. A comparison of our results from low temperature experiments with published values from high temperatures makes obvious that the dependence of diffusivity on temperature cannot be described by an Arrhenius relation across the glass transition for the basic compositions that were studied by us. Thus, a simple extrapolation between the two temperature domains is not applicable, which must be considered in low temperature applications. In contrast, our results for cpx confirmed that diffusivities from high temperatures (> 600 °C) can be extrapolated to low temperatures. The newly obtained data for cpx was used in a non-isothermal diffusion model that simulates ascent of a crystal along representative pressure-temperature trajectories with variable boundary conditions. Our results provide information on when re-equilibration due to diffusion at low temperatures of samples from high temperature settings (e.g. the mantle of the Earth) must be considered - for example, when ascent chronometers are applied, or H contents are interpreted as signature of a mantle source.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung