Published in

Elsevier, Earth and Planetary Science Letters, 3-4(257), p. 609-621

DOI: 10.1016/j.epsl.2007.03.023

Links

Tools

Export citation

Search in Google Scholar

High-temperature lithium isotope fractionation: Insights from lithium isotope diffusion in magmatic systems

Journal article published in 2007 by Ian J. Parkinson, Samantha J. Hammond ORCID, Rachael H. James, Nick W. Rogers
This paper was not found in any repository, but could be made available legally by the author.
This paper was not found in any repository, but could be made available legally by the author.

Full text: Unavailable

Green circle
Preprint: archiving allowed
Red circle
Postprint: archiving forbidden
Red circle
Published version: archiving forbidden
Data provided by SHERPA/RoMEO

Abstract

Ion-microprobe analyses of the Li concentration and Li isotopic composition of zoned clinopyroxene and olivine phenocrysts from within primitive arc lavas from the New Georgia Group in the Solomon Islands reveal that both Li and δ7Li vary widely from rim to core. The Li content of the rims is between 2 and 8 times that of the cores whereas Li isotope profiles are characterised by a zone with low δ7Li (as low as − 20‰) and cores with δ7Li values of between − 4 and + 8‰; these over-print macroscopic major element zoning. With time, the low δ7Li zone broadens and migrates towards the centre of the crystal and the Li concentration gradient is reduced. These data are consistent with preferential diffusion of 6Li into the grain from a Li-enriched rim with 6Li diffusing ∼ 3% faster than 7Li. The profiles of δ7Li and Li concentration can be reproduced by numerical modelling which confirms that the size of the δ7Li trough is a function of the Li concentration gradient and the fractional difference in the diffusion rates of 6Li and 7Li. Both open and closed system models predict that a zone with low δ7Li will migrate through the mineral grain with time, eventually relaxing back to a flat profile. Modelling of Fe-Mg diffusion in olivine suggests that the crystals have a residence time of 13–150 days, which is in accordance with the observed Li isotope profiles. This allows us to calibrate the rate of Li diffusion, which is 4–8 times slower in olivine and 20–30 times faster in clinopyroxene than Fe–Mg diffusion in olivine. The high speed of Li diffusion means that the δ7Li values of minerals that interact with Li-rich melts can rapidly decrease. Therefore, porphyritic lavas are unlikely to be suitable for Li isotope studies of mantle processes and it may also explain why olivines generally have higher δ7Li than co-existing pyroxenes in some mantle samples. Modification of Li isotope ratios, by interaction with the host lava, may occur in mantle xenoliths during transport to the Earth's surface in only a few days. Conversely, melts ascending through the mantle will rapidly exchange Li and this may erase the pristine δ7Li information that the melt carries. This may explain why many subduction zone lavas do not have an obvious slab signature. This study demonstrates that Li diffusion can overprint primary mineral compositions on very short timescales. This means that careful investigation of coexisting minerals is required, but it may also provide valuable information about the timescales of short duration events.