Research paperMelt inclusion Pb-isotope analysis by LA–MC-ICPMS: Assessment of analytical performance and application to OIB genesis
Highlights
► Pb isotopes in olivine-hosted melt inclusions, including 204Pb, have been measured by laser ablation for the first time. ► These laser ablation results are far more restricted than previous results. ► Levels of Pb isotopic variability must be geological rather than analytical. ► Pb isotopic systematics in the Pitcairn Seamounts can be attributed to variation within the plume.
Introduction
Silicate melt inclusions (hereafter ‘melt inclusions’) are small portions of melt, trapped by crystals growing within magmas (Roedder, 1979). It is commonly accepted that once trapped, the incompatible trace element signature of the melt is effectively preserved, being isolated from subsequent chemical evolution, and thus provides a detailed record of early magmatic history that may not always be evident from bulk-rock studies. As a result, melt inclusions may have great significance in the interpretation of magmatic processes and magma evolution.
It has also been suggested that melt inclusions in high-magnesian olivine phenocrysts which show distinctly different major and trace element concentrations to their host rocks may actually reflect grain-scale phenomena (Danyushevsky et al., 2004), where hot, primitive melts react within the magmatic system with more evolved crystal phases, resulting in the trapping of a melt which would not necessarily be representative of the macro system. This then raises the critical question as to whether those melt inclusions in olivine phenocrysts, which are characterised by a significant range of compositions, extending well outside the range observed in host lavas, are truly representative of geologically significant magma bodies, or are simply artefacts of localised grain-scale phenomena.
The ability to determine isotopic compositions in melt inclusions offers the prospect of some resolution to this debate since isotope ratios may be insensitive to many of the localised melting reactions that would otherwise affect major and/or trace element concentrations. The limited research in this area to date, however, has produced conflicting observations. Published Pb isotope ratios (Saal et al., 1998, Yurimoto et al., 2004, Saal et al., 2005) measured in olivine-hosted melt inclusions employing secondary ion mass spectrometry (SIMS) analytical techniques suggest that they do in fact record large compositional variations not seen in their hosts. For example, nearly 50% of the Mangaia (Cook Islands, S. Pacific) melt inclusions reported by Saal et al. (2005) exhibit compositions that are not observed in erupted lavas (Fig. 1). In contrast, however, Sr isotope measurements of melt inclusions from Samoa in the S. Pacific (Jackson and Hart, 2006) and a limited number (n = 8) of more recent measurements of Pb isotope determinations in melt inclusions from Erebus volcano, Antarctica (Sims et al., 2008), appear to be entirely consistent with their hosts, leaving considerable ambiguity in the interpretation of melt inclusion isotope data.
In this contribution, we provide the first laser ablation MC-ICPMS Pb isotope data for olivine-hosted melt inclusions, combined with an in-depth investigation of analytical performance based upon analyses of both glass reference materials and a sample from Tonga which, based upon previous studies, is assumed to contain a single population of melt inclusions. New measurements are then provided for the end-member OIB localities of Mangaia (‘HIMU’) and the Pitcairn Seamounts (‘EM-I’) (Fig. 2; Zindler and Hart, 1986) providing a method-independent evaluation of the published SIMS data, much of which is based on samples from these same locations. One of the advantages of laser ablation is that it is a relatively rapid technique requiring little sample preparation, and in this study we present 90 new melt inclusion Pb isotope analyses, representing roughly one third of the existing global database (Saal et al., 1998, Kobayashi et al., 2004, Yurimoto et al., 2004, Saal et al., 2005, Maclennan, 2008, Sims et al., 2008). Laser ablation techniques typically consume far more sample than SIMS-related techniques and thus sampling volumes may be more representative of whole inclusions. Furthermore, the larger analyte volume also permits measurement of the critical non-radiogenic isotope 204Pb, allowing comparison with conventional Pb isotope bulk-rock data sets, in addition to the 207Pb/206Pb and 208Pb/206Pb ratios. Our new data are compared with existing literature values and the geological implications discussed.
Section snippets
North Tongan boninite
Our ‘reference sample’, used primarily to characterise the limitations of the analytical method, comes from the Tonga subduction system in the S.W. Pacific. Boninites from the northern termination of the Tofua Arc have been well studied in terms of their petrology, geochemistry and mineralogy, and are one of very few reported occurrences of boninites from an active arc setting (Sharaskin et al., 1983, Falloon and Green, 1986, Falloon et al., 1987, Falloon et al., 1989, Sobolev and Danyushevsky,
Sample preparation
Initial preparation for Tongan sample 16-26-2 was conducted at CODES and differs slightly from the preparation methods employed at the University of Melbourne for the Mangaia and Pitcairn samples. The Tongan sample was hand-crushed, and individual olivine phenocrysts were handpicked and mounted in an epoxy grain mount. Grains containing primary melt inclusions > 100 μm in diameter were extracted and inclusions were homogenised to glass using a low-inertia microscope-mounted heating stage (Sobolev
Analytical accuracy
The only robust method of estimating analytical accuracy for melt inclusion analysis, albeit under near ideal conditions, is through the analysis of fused glass reference materials with similar Pb contents to those expected in the melt inclusions under study. Two such materials were run concurrently with our melt inclusion samples and the data reduced as unknowns: these were ATHO, a fused Icelandic rhyolite with ~ 6 μg.g−1 Pb prepared by the Max Planck Institute (bulk Pb-isotope data presented by
Conclusions
A new comprehensive dataset of melt inclusion Pb-isotope analyses from key OIB samples, obtained by laser ablation ICPMS, is presented. When compared to homogeneous glass standards and well-characterised homogeneous inclusion populations, the melt inclusion compositions measured in this study show more variation than can be attributable solely to analytical variation. The Pb isotope results reported, however, show distinctly less variation than previous measurements by other analytical
Acknowledgements
This research was funded by the Australian Research Council through the ARC Centre of Excellence programme. JW's participation in the SO65 cruise to the Pitcairn Seamounts was supported by a German Research Ministry grant to P. Stoffers, University of Kiel. This study was also supported by the “Centre of Excellence for the 21st Century in Japan” programme of the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT) to E.N.
Alberto Saal and John Maclennan are thanked for
References (55)
- et al.
Morphostructural study and type of volcanism of submarine volcanoes over the Pitcairn hot spot in the South Pacific
Tectonophysics
(1992) - et al.
Temporal evolution and geochemical variability of the South Pacific superplume activity
Earth and Planetary Science Letters
(2006) - et al.
Melting of enriched mantle beneath Pitcairn Seamounts: unusual U–Th–Ra systematics provide insights into melt extraction processes
Earth and Planetary Science Letters
(2009) - et al.
Pb isotopic heterogeneity in basaltic phenocrysts
Geochimica et Cosmochimica Acta
(2004) - et al.
Contrasting old and young volcanism in Rurutu Island, Austral chain
Chemical Geology
(1997) - et al.
North Tongan high-Ca boninite petrogenesis: the role of Samoan plume and subduction zone–transform fault transition
Journal of Geodynamics
(1995) The role of sediment recycling in EM-1 inferred from Os, Pb, Hf, Nd, and Sr isotope and trace element systematics of the Pitcairn hotspot
Earth and Planetary Science Letters
(2002)- et al.
Glass inclusions in magnesian olivine phenocrysts from Tonga: evidence for highly refractory parental magmas in the Tongan arc
Earth and Planetary Science Letters
(1986) - et al.
Precise lead isotope measurements by the double spike technique — a reconsideration
Geochimica Et Cosmochimica Acta
(1985) - et al.
Re–Os systematics of HIMU and EMII oceanic island basalts from the south Pacific Ocean
Earth and Planetary Science Letters
(1993)