- Iron oxidation in (Mg,Fe)0: Calibration of the Flank method on synthetic samples and applications on natural inclusions from lower mantle diamonds (2009)
- (Mg,Fe)O ferropericlase is the most common mineral found in diamonds originating in the lower mantle (more than 50% of occurrences). It is well known that the Fe3+ concentration in (Mg,Fe)O is sensitive to oxygen fugacity, even at high pressures. Therefore, the determination of Fe3+/Fetot in such inclusions provides a direct method for investigating lower mantle redox conditions during diamond formation. The goal of the present research is to calibrate the “flank method” by electron microprobe using synthetic (Mg,Fe)O, and then apply the method to determine in situ Fe3+/Fetot in ferropericlase inclusions from lower mantle diamonds. Up to now a calibration of the flank method is available only for garnets. Initially, the flank method was calibrated for garnets to test the reproducibility of the method on the Jeol XA-8200 electron microprobe in use at Bayerisches Geoinstitut. Results showed that for garnets a new calibration curve needs to be established at each working session. Then the flank method was calibrated for the Jeol XA-8200 electron microprobe in use at Bayerisches Geoinstitut for a homogeneous set of (Mg,Fe)O ferropericlase crystals over a wide range of composition (xFe = 2 to 60 at.%) and Fe3+/Fetot (1 to 15%). Samples were obtained by performing high pressure high temperature experiments in a multi anvil apparatus. In order to avoid compositional effects on flank method measurements, the high sample homogeneity was essential. Moreover, the determination of the Fe3+/Fetot ratio needed to be extremely accurate. For this purpose, a more accurate procedure for fitting the Mössbauer spectra of the final set of synthetic (Mg,Fe)O was adopted. The calibration curve determined is Fe2+ = 46.238 + 8.161 * ln (Fetot) - 137.01 * (Lbeta/Lalpha) + 85.57 * (Lbeta/Lalpha)2, for a Fe compositional range between 3 and 47 wt. %. A comparison of Fe3+/Fetot determined by flank method and values determined earlier by Mössbauer spectroscopy shows that results are generally consistent between the two different methods within the experimental errors. In contrast with garnet, the calibration curve established for ferropericlase does not need to be recalibrated at each microprobe session. Therefore, the calibration curve can be considered universal for the electron microprobe in use if the spectrometer adjustments remain identical with time. To explore applications of the flank method, a set of (Mg,Fe)O samples from diffusion studies was also investigated. Three (Mg,Fe)O crystals were measured by electron microprobe in order to test the sensitivity and accuracy of the flank method for small variations of bulk (Fetot)(wt%) as well as to measure Fe3+/Fetot along diffusion profiles. In the present work it is demonstrated how the flank method can be a powerful tool to measure small variations in Fe3+ content, with a spatial resolution of only few microns (2-3 µm) and a lower detection limit of Fetot of 3 wt%. Moreover, the measurement of Fe3+ content on the micron scale enables the study of the variation of oxygen fugacity conditions along diffusion gradients. A set of (Mg,Fe)O ferropericlase inclusions from ultra deep diamonds selected worldwide were analyzed by the flank method. The data set consists of eighteen (Mg,Fe)O ferropericlase samples from Juina, Brazil, Machado River, Brazil, and Ororoo, Australia. Inclusions are between 10 and 50 µm in size, therefore they are suitable to perform flank method measurements to determine Fe3+/Fetot. For the first time Fe3+/Fetot ratios were measured directly at the electron microprobe on inclusions of less than 50 µm in size. Results for the (Mg,Fe)O inclusions show good agreement with the theoretical trend described by the synthetic samples, which confirms high phase homogeneity for most of the samples. Flank method measurements show a large range of Fe3+/Fetot values for (Mg,Fe)O inclusions, which implies a large range of oxygen fugacities based on charge balance calculations. This large range of oxygen fugacities is similar to results for a suite of much larger inclusions from Kankan, Guinea, and São Luiz, Brazil, that were studied using Mössbauer spectroscopy. The variation of oxygen fugacity seems to be correlated to the geographical distribution of the inclusions studied, showing a redox gradient with more reducing conditions at Kankan, Guinea, and São Luiz, Brazil, and more oxidized in the case of Juina and Machado River, Brazil, and Eurelia, Australia. Such a correlation may be linked to the proto-pacific subduction mechanism, and the different ages combined with the geographic variation may indicate a difference in depth correlating with the large redox variation. Inclusions recovered from the same host diamond from Eurelia shows a strong redox gradient, which suggests a drastic change in the oxygen fugacity conditions during diamond growth. In order to provide information on the mechanisms able to control the redox conditions at lower mantle depths, a multi disciplinary study is suggested for further work.