高温下非传统稳定同位素分馏

过去十几年来,非传统稳定同位素地球化学在高温地质过程的研究中取得了的重大进展。多接收诱导耦合等离子质谱(MC-ICP-MS)的应用引发了稳定同位素分析方法的重大突破,使得精确测定重元素的同位素比值成为可能。本文总结了以Li、Fe和Mg同位素为代表的非传统稳定同位素在岩石地球化学研究中的应用。Li同位素目前被广泛地用于地幔地球化学、俯冲带物质再循环和变质作用的研究中,可以用来示踪岩浆的源区性质和扩散等动力学过程。不同价态的Fe在矿物熔体相之间的分配可以产生Fe同位素分馏,可以发生在地幔交代、部分熔融、分离结晶等过程中。岩浆岩的Mg同位素则大致反映其源区的特征,地幔的Mg同位素组成比较均一,这为研究低温地球化学过程中Mg同位素的分馏提供一个均一的背景。此外,Cl,Si,Cu,Ca,U等等同位素体系也具有广阔的应用前景。对同位素分馏机制的实验研究和理论模拟为理解非传统稳定同位素数据提供了必要的指导。实验表明,高温下具有不同的迁移速度的轻、重同位素可以产生显著的动力学同位素分馏,这一分馏可以在化学扩散、蒸发和凝华等过程中发生;同位素在矿物和熔体以及流体相中化学环境的差异使得不同相之间可以发生平衡分馏。而最近的硅酸盐岩浆的热扩散和热迁移实验则揭示了一种"新"的岩浆分异和同位素分馏机制。沿着温度梯度,硅酸盐岩浆可以发生显著的元素和同位素分异,湿的安山岩可以通过这种方式演变成花岗质成分,因此这个过程可能对陆壳的产生和演化有重大影响。如果温度梯度在岩浆作用中能长期存在,热扩散就可以产生稳定同位素的分馏,这一机制有别于传统的平衡和动力学同位素分馏。 而多个稳定同位素体系的正相关关系是示踪热迁移过程的最有力证据。在热扩散过程中,流体承载的物质的浓度和它的索瑞系数有关。但是这个系数对体系的很多参数非常敏感,变化极大,因此对热扩散效应的研究产生极大的困难。对热扩散实验的镁、钙和铁同位素测量表明,同位素比值的变化与体系的化学组成以及总温度无关,只和温度变化的幅度有关,这意味着即使元素的索瑞系数变化多端,某一元素的同位素之间的索瑞系数的差别总为常数。这一发现有助于简化对热扩散和索瑞系数这一基础物理问题的研究 。

Non-traditional stable isotope fractionation at high temperatures

The last ten years have seen big progress and wide applications of a novel field, non-traditional stable isotope (NTSI) geochemistry, to high temperature geo-science studies. Invention of multi-collector-inductively coupled plasma-mass spectrometry (MC-ICP-MS) led to the big breakthrough of analytical methods for heavy stable isotopes. This contribution summarizes Li, Fe, and Mg isotope studies on igneous rocks and minerals, as representative of NTSI geochemistry.Li isotopes have been widely applied to the studies of mantle geochemistry, recycling of subducted materials, and metamorphism to constrain the source of magma and kinetic diffusion process. Fe isotope fractionation is related to partitioning of multi-valent Fe between Fe-bearing phases, which can occur in the course of mantle metasomatism, partial melting, and fractional crystallization. Mg isotopic compositions of igneous rocks most likely reflect the source signatures.Variation of Mg isotopic ratios of mantle peridotites is trivial and this provides a homogenous background for Mg isotope fractionation in low temperature processes. Furthermore, Cl, Si, Cu, Ca, and U isotopes are also promising in the future geochemical studies . Experimental studies and theoretical simulation for the mechanisms of isotope fractionation provide important guidances for understanding the NTIS data. Experimental studies show that light and heavy isotopes have different migration velocity at high temperature processes such as chemical diffusion, evaporation, and desublimation, which could produce significant kinetic isotope fractionation. Equilibrium isotopic fractionation could occur among mineral, melt, and fluid when chemical environment of the isotopes are different between the phases. Recent thermal diffusion and migration experiments on silicate material reveal a "new" mechanism of magma differentiation and isotope fractionation. Along a temperature gradient in silicate magma, large elemental variation and isotopic fractionation can occur, by which a wet andesite can even be differentiated to granite. This suggests that thermal migration could be important for continental crustal formation and evolution. If temperature gradient exists long enough during magma differentiation, thermal diffusion can produce significant stable isotope fractionation, which is contrast to the mechanism of traditional kinetic and equilibrium isotope fractionations. Such process can be fingerprinted by positive correlations among multi-stable isotopic systems. Due to thermal diffusion, concentration of material loaded or dissolved in the fluid is a function of Soret coefficient (S_T). However, because S_T is highly variable and sensitive to lots of factors, the basic physics of thermal diffusion is still poorly understood. As shown by Mg, Ca, and Fe isotope measurement of thermal diffusion experiments, isotope fractionation driven by temperature gradient is independent to the bulk composition and temperature of the system, suggesting that the difference of S_T between two isotopes of the same element can be considered as a constant. This can simplify and help the studies on thermal diffusion and S_T.