Lithium isotope fractionation in magmatic systems : constraints from in situ δ7Li determinations on magmatic minerals by femtosecond-laser ablation-MC-ICP-MS

• Main Author: Steinmann, Lena
• Format: Article in Journal/Newspaper
• Language: English
• Published: Hannover : Institutionelles Repositorium der Leibniz Universität Hannover 2020
• Subjects: Lithium isotope fractionation; olivine; matrix-dependent isotope fractionation in the plasma; diffusion modeling; multi-stage magma evolution; magmatic degassing; Massif Central France; Lithium-Isotopenfraktionierung; Olivin; fs-LA-MC-ICP-MS; matrixabhängige Isotopenfraktionierung im Plasma; Diffusionsmodellierung; mehrstufige Magmenevolution; magmatische Entgasung; Massif Central Frankreich; Dewey Decimal Classification500 | Naturwissenschaften550 | Geowissenschaften; Kamchatka Peninsula; Faraday; Dewey; Kamchatka
• Online Access: https://dx.doi.org/10.15488/9918; https://www.repo.uni-hannover.de/handle/123456789/9976

Investigations on the fractionation of stable metal isotopes to characterize mass flux at high temperatures have been proven to be a powerful tool during the past years. In this study, high precision in situ analyses on Li isotope ratios were performed on reference glasses and natural olivines at low concentration levels by femtosecond-laser ablation multi collector inductively coupled plasma mass spectrometry (fs-LA-MC-ICP-MS) in order to investigate the fractionation of Li isotopes during magma evolution. The analytical technique was tested by analyzing a series of reference glasses in situ and cross calibrating them with published values acquired by other methods (solution nebulization MC-ICP-MS, SIMS and TIMS). The results of this methodical investigation showed that operating the plasma at relatively cool conditions (900 W) largely suppresses matrix-dependent isotope effects in the plasma. In order to achieve the best precision for concentrations ranging from 2 to 10 µg/g a detector combination of an ion counter for the determination of 6Li and a faraday cup equipped with a 1013 Ω amplifier for 7Li was applied. Precise and accurate measurements of δ7Li with ~2 ‰ (2 σ) analytical uncertainty were performed on reference glass T1-G (δ7LiT1 G = 1.6-2.4 ‰). The fractionation of Li isotopes on the outcrop and mineral scale were investigated in this study. Bulk Li isotope analyses of an outcrop over a length of ~50 m in the French Massif Central showed that on the outcrop scale the isotopic composition varied between +2.1 and +3.3 ±2.0 ‰. The measured δ7Li-values coincide with the range of unaltered volcanic whole-rock suites worldwide (+2.0 to +5.0 ‰) and the value determined for the bulk silicate Earth (+3.5 to +4.0 ‰). A systems analysis was performed in order to determine melt reservoirs for two locations of distinct geotectonic settings (ocean intra-plate (Tenerife, Canary Islands, Spain) and volcanic island arc (Kluchevskoy volcano, Kamchatka peninsula, Russia)). Reservoirs of varying primitive grade in a volcanic plumbing system were determined and the passage way of crystals could be retraced by forsterite and Li contents. Diffusion of Li was modeled to be 1.4–2.5 times faster than that of the Mg-Fe diffusion couple, which is significantly less than the diffusivity determined in experimental studies. Chemically and isotopically zoned olivines from a continental intra-plate setting (Massif Central, France) were analyzed regarding their Li isotope composition along profiles from rim to core. Variations in the Mg-Fe and Li isotopic composition revealed a diffusive origin of the zoning, and Li gives insight into a second diffusive event, which remains hidden in Mg-Fe. Profiles of the first diffusive event were modeled with fixed time scales, based on Mg-Fe inter-diffusion, acquired by Oeser et al. (2015), for the residence time in a magma reservoir. The model again resulted in lower diffusion coefficients for Li diffusion in olivine, than those determined in experimental studies. A second, relatively short-lived, diffusive re-equilibration episode was assumed to be caused by the degassing of the magma and concomitant Li decrease in the melt with no effect on Mg-Fe distribution.