Search for: All records

Mantle heterogeneity has a first-order control of the petrological and geochemical differences of erupted mafic lavas worldwide. Whether this heterogeneity reflects only chemical variations or also lithological heterogeneity in source regions is debated. Because of their contrasted partitioning behaviors between mantle phases, First Row Transition Elements (FRTEs) are considered as potential lithological tracers. Using a combination of published data on natural and experimental samples and new high current microprobe analyses on a variety of pyroxenite samples, we investigated the parameters that control FRTE exchange coefficients (Kd) between common mantle minerals and performed inverse modeling to test if FRTE ratios from oceanic basalt compositions can be used to solve for modal proportions in their mantle source. We applied the Kd determined from mantle lithologies in this study, along with experimental melt-mineral partitioning coefficients and a simplified melting model, on two basalt suites selected for their contrasted Mn/Fe and Zn/Fe ratios. Our results show that a same FRTE ratio can be explained by a range of modal proportions in the source. However, when combined, FRTE ratios become a powerful tool to constrain the nature of the source. 

Mantle heterogeneity has a first-order control on the petrological and geochemical differences of erupted mafic lavas across the globe. It is debated whether this heterogeneity reflects only chemical variability or also lithological differences in source regions. Because of their various partitioning behaviors between mantle minerals, First Row Transition Elements (FRTEs) have been identified as potential lithological tracers. Here, we investigate the various parameters that control FRTE partitioning between common mantle phases through a comparison of partition coefficients calculated from natural pyroxenites obtained from the Earthchem database with previous partitioning experiments and new electron microprobe analyses. Using naturally occurring pyroxenites from alpine massifs and xenoliths provides the opportunity to explore the behavior of FRTEs on a much larger range of compositions and temperatures than covered by experimental studies. Our preliminary results show that natural partition coefficients for Fe and Mn depend on temperature and vary distinctly between lithologies. The effect of composition, however, is difficult to resolve and will require further inspection. Natural exchange coefficients, or Kd’s (mineral/mineral) for Mn/Fe, largely match previous experimental data across peridotite and pyroxenite compositions for garnet/clinopyroxene(cpx), orthopyroxene/cpx, and olivine/cpx. However, natural samples often present evidence of chemical disequilibrium and/or secondary alteration which can significantly increase the scatter in analyses. Importantly, despite the larger uncertainty on the natural Kd’s than on experimental ones, natural exchange coefficients show distinct values between the various pairs of minerals. These distinctions, and the fact that Kd’s do not seem to be influenced by temperature, make the bulk Mn/Fe ratio in lavas a good lithological tracer, supporting previous claims. Hence, we show that natural compositions can be used to expand trends in FRTE distribution behavior across a wider range of temperatures (500-1500°C) and compositions than determined previously by experiments alone. 

The quantification of the mantle heterogeneity and its contribution to magma genesis are crucial for our understanding of the nature of Earth’s geochemical reservoirs and recycling processes. Today, the source of basalts is often envisioned as a heterogeneous mantle that comprises a range of lithological heterogeneities, especially pyroxenites, introduced into the mantle by various geodynamic and magmatic processes. Different chemical parameters have been proposed to trace evidence for the contribution of lithological heterogeneities during magma genesis (e.g., major elements1-2, major element ratios1,3 or logratios4, first row transition element concentrations5-6, iron isotopes7) and several empirical models for partial melting of a heterogeneous mantle have been developed in an attempt to quantify the proportion of pyroxenites in the mantle source of magmas8-11. However, the large range of compositions covered by the potential lithologies present in the mantle makes it challenging to determine a unique proxy for their contributions in the magmas2,12. Additionally, these contributions are controlled by various factors, such as the abundances of the different lithologies in the mantle, their respective melting behaviors (solidus temperatures and melt productivities) and the thermal and melting regimes of the mantle. Finally, interpretations from the magma compositions are complicated by the potential presence of volatiles in the mantle source and/or by modifications experienced by the magmas during their journey through the mantle8,12. This keynote presentation will present examples of challenges that can rise when we try to quantify the nature and abundance of the mantle heterogeneity, the efforts that have been published by various authors, and a few potential research directions that could bring prospects for success. 1- Hauri (1996), Nature 382, 2- Lambart et al. (2013), Lithos 160-161, 3- Hirschmann et al. (2003), Geology 31, 4- Yang et al. (2019), JGR-Solid Earth 124, 5- Sobolev et al (2007), Science 316, 6- Sobolev et al. (2008), Science 321, 7- Williams and Bizimis, EPSL 404, 8- Mallik and Dasgupta (2014), GGG 15, 9- Kimura and Kawabata (2015), GGG 16, 10- Lambart et al. (2016), JGR-Solid Earth 121, 11- Brown and Lesher (2016), GGG 17, 12- Mallik et al. (in press), in: Konter J., Ballmer M, Cottaar S, & Marquardt H. (Eds. ), AGU monograph. 

Natural forsterite(Fo)-rich olivines represent the major constituent of the Earth's upper mantle. Collected at the surface in mantle xenoliths, they are commonly used in scientific research. Many electron microprobe labs use the San Carlos standard USNM111312/44 [1], provided by the Smithonian Institution, for calibrations. Non-USNMdistributed crystals of San Carlos olivine are also often used as starting material in experimental studies [e.g., 2, 3]. However, the potential inherent chemical variability of starting materials can affect results and their scientific interpretations. Hence, it is important to characterize the full chemical variability of the San Carlos olivine. Fournelle [4] showed that the USNM material shows limited variability (Fo89.6 to Fo90.5), but that non-standard San Carlos olivines can be significantly more variable, with Fo contents ranging from 87 to 92%. Following these results, we report new major and trace element analyses on grains (0.5 mm - >5mm) of non-USNM San Carlos olivine and compare them with analyses on USNM San Carlos standard compositions. We also investigate the presence of potential grain-scale chemical variations by looking at composition profiles on large (> 5mm) grains. Observed major-element variations (Fo88.4 to Fo91.4) are consistent with Fournelle’s results [4]. Additionally, we show that minor and trace element concentrations present significant and contrasted variations between grains (e.g., 17 % Ni, 28 % Mn, 44% V, 69 % Al, 285 % P, relative). At the scale of the individual grain, however, San Carlos olivines appear relatively homogeneous with no systematic core-rim variations. Results and implications for the use of this material in experimental studies and for interpretations of the petrogenetic processes will be discussed.