Combined chemical and high-precision isotope analysis of silicate melt inclusions – Advancing the understanding of Earth’s heterogeneous mantle

对MI的研究提供了对地幔成分的深入了解，地幔成分隐藏在载体矿物或整个岩石中。早结晶橄榄石中的熔体与与大块岩石平衡的熔体相比，其化学成分差异较大，变化较大。在提议的项目中，将分析玄武岩熔岩中原始硅酸盐（橄榄石、辉石）的化学和同位素组成。最近对熔体包裹体同位素组成的研究表明，包裹体的同位素范围几乎超过了喷发玄武岩的一个数量级，即地幔中同位素非均质性的程度比以前认为的要大得多。然而，到目前为止，还没有来自单个熔体包裹体的一致的主要、微量元素和同位素数据集。然而，将熔体包裹体的成分与喷发玄武岩的成分进行比较，对地幔的实际非均质性以及部分熔融和随后的熔体演化如何影响喷发熔体的成分提供了重要的见解。本项目选择的样本来自与羽流相关的海洋岛屿，如冰岛、夏威夷、特里斯坦达库尼亚岛，以及法罗群岛、亚速尔群岛、圣赫勒拿岛、高夫和库克南岛，这些岛屿几乎代表了所有已知的同位素谱。到目前为止，熔体包裹体的同位素组成都是用原位法测定的；用SIMS或LA-ICP-MS测定。由于通常10-100微米大的熔体包裹体的离子产率很低，并且有相当大的等压干涉校正，原位技术的测量误差通常比化学分离样品溶液的TIMS和MC-ICP-MS测量误差大10-100倍，这两种方法不需要（或很少）等压干涉校正。例如，Sobolev等人（2011）报告了使用单收集器LA-ICP-MS分析熔体包裹体中Sr同位素的误差为0.0005 (S.D.)，即700 ppm。或者，最新的多收集器质谱仪，如Thermo Neptune Plus MC-ICP-MS，提供高测量灵敏度（2000V / ppm Sr），可以对几纳克材料进行高精度分析，这是典型熔体内含物中Sr含量的要求。利用高灵敏度，小于5 ng的量就足以在ppm水平上进行Sr同位素分析，与原位方法相比，这是一个数量级以上的改进。MI同位素测量精度的提高将提供地幔（同位素）非均质性的高分辨率图像。结合化学信息，如主要元素和微量元素，将获得关于熔融和结晶过程的新见解，这对于更好地了解地球地幔组成至关重要。

Studies of MI provide insights into the composition of the Earth’s mantle, which are concealed in the carrier minerals or the whole rock. Trapped melts in early-crystallized olivines have significantly different and more variable chemical compositions than melts in equilibrium with the bulk rock. During the proposed project, MI in primitive silicates (olivine, pyroxene) from basaltic lavas will be analysed for their chemical and isotopic composition. Recent studies on the isotopic composition of melt inclusions have shown that the isotopic range of the inclusions exceeds that of the erupted basalts by almost an order of magnitude, i.e. the extent of isotopic heterogeneity in the mantle is much larger than previously thought. Yet, to date no consistent major-, trace element and isotope datasets from individual melt inclusions are available. Comparisons of the composition of melt inclusions and the erupted basalts, however, provide important insights about the actual heterogeneity in the mantle and about how partial melting and subsequent melt evolution influence the composition of the erupted melts. The samples selected for this project are from plume-related ocean islands, such as Iceland, Hawaii, Tristan da Cunha, but also the Faroe Islands, the Azores archipelago, St. Helena, Gough, and the Cook Australs that are represnetative of almost the entire known isotopic spectrum. So far, the isotopic composition of melt inclusions was determined exclusively by means of in-situ methods; either by SIMS or by LA-ICP-MS. Owing to the low ion yield of typically 10-100 micrometre large melt inclusions and the considerable isobaric interference corrections, the measurement errors of the in-situ techniques are often 10-100 times larger than that of TIMS and MC-ICP-MS measurements of chemically separated sample solutions, which require no (or minimal) isobaric interference corrections. For example, Sobolev et al. (2011) report an error of 0.0005 (S.D.), i.e., 700 ppm for Sr isotope analysis in melt inclusions using single collector LA-ICP-MS. Alternatively, the latest multi-collector mass spectrometers such as the Thermo Neptune Plus MC-ICP-MS offer high measurement sensitivity (2000V / ppm Sr) which enables high-precision analyses of a few nanogram material, as is required by the Sr content in typical melt inclusions. Making use of the high sensitivity, quantities of less than 5 ng are sufficient for Sr isotope analyses to be carried out on the ppm level, which is an improvement of more than one order of magnitude compared to in-situ methods. The improved precision of MI isotope measurements will provide a high-resolution image of the (isotopic) heterogeneity of the mantle. In combination with chemical information, such as major and trace elements, new insights into melting and crystallisation processes will be obtained, which are crucial for a better understanding of Earth’s mantle composition.