The Earth is unique among the planets and moons of the solar system in having a highly differentiated continental crust. The formation of this crust required a complex process including several stages of melting of rocks and crystallization of magmas. Models for the differentiation of the Earth are guided by the abundances of elements in crustal and mantle rocks. The general assumption is that minerals and melts are in complete equilibrium with respect to their element concentrations in all of these melting and crystallization processes. Equilibrium concentration ratios between minerals and melts have been determined experimentally in the laboratory and are conventionally used to model crust-mantle differentiation. The element pair niobium (Nb) and tantalum (Ta) has been identified as critical for the distinction of different melting regimes involved in crustal differentiation, but published equilibrium models have largely failed to reproduce the observed Nb-Ta distribution patterns.
In a series of experiments in preparation for this proposal, Dr. Marschall and his collaborators have investigated the diffusion of Nb and Ta in rutile (titanium dioxide). Rutile is the dominant mineral host of Nb and Ta in many rocks. Their experiments unequivocally demonstrate that Nb diffuses significantly faster in rutile than Ta, and they demonstrate that Nb and Ta may be separated by diffusion during crustal melting processes. Taking the range of grain sizes of rutile in natural rocks and the temperatures and time scales involved in the melting of crustal rocks, it is demonstrated that Nb-Ta equilibrium cannot be expected. The findings suggest that the low Nb/Ta ratio documented for the continental crust is the results of partial?as opposed to complete?equilibration of rutile and melt, and that the assumption of equilibrium melting has to be abandoned for the element pair Nb-Ta in the processes of crustal differentiation. The goal of this study is a detailed determination of the diffusion of Nb and Ta in rutile. This data will be applicable to geochemical modeling and for the determination of time scales of geological processes operating deep in the Earth crust. The study sets out to characterize Nb and Ta diffusion in rutile as a function of various parameters, such as temperature, and the minor element composition of rutile. This will allow the identification of the mechanisms of incorporation and of diffusion of Nb and Ta in rutile, and it will quantify the parameter space for equilibrium vs. disequilibrium melting in natural systems. The latter will be employed to develop and refine the geochemical models for crustal differentiation.
Apart from the growing interest of the geological community in the mineral rutile, there is an even greater demand by material scientists for quantitative data on trace-element incorporation and diffusion mechanisms in rutile. Technical applications of rutile include devices for the photocatalytical hydrolysis of water and photovoltaic installations, which are based on the semiconducting properties of rutile. However, its performance in these appliances is still hindered by the low electrical conductivity of pure rutile, and the difficulties of reproducing the desired properties. Doping of rutile with Nb can turn it into a semiconductor or even a quasi-metallic conductor. Nevertheless, for an industrial application of Nb-doped rutile semiconductors, it is essential to determine the diffusivities of Nb in rutile as a function of temperature, oxygen pressure and concentration of impurities. These data are needed for the technical processing in order to produce homogeneous materials with a well-defined composition and the desired properties. This study sets out to provide these data.
