Diffusion Kinetics of Selected Trivalent Ions and Hf in Garnet: Experimental Studies and Applications

扩散是一个基本的原子过程，它是固体在相对高温下各种行为的基础，因此是许多科学和工程领域的主要兴趣课题。虽然姗姗来迟，但现在对造岩矿物的扩散特性的研究产生了很大的兴趣，因为对扩散控制特性的建模对山带演化的时间尺度、构造板块的俯冲速度、地球内部的动力学过程（如熔融关系和岩浆产生、对流）、火山爆发前岩浆房中熔体的停留时间尺度，具有预测火山活动的潜力等。本研究的重点是(a)实验确定天然石榴石中选定的阳离子（Sm3+、Lu3+、Y3+和Hf4+）的扩散动力学性质，作为影响扩散过程的重要物理和化学因素的函数，以及(b)将数据应用于大谱地质过程时间尺度的定量建模。将这些数据与该团队对石榴石中二价（2+）阳离子扩散特性的早期研究结果相结合，将是朝着开发固体扩散预测模型迈出的重要一步，这对固体扩散的一般领域具有重要意义。根据样品和其他细节，实验诱导的扩散谱将通过离子探针的SIMS（二次离子质谱）深度谱分析，纳米SIMS的步进扫描或我们最近获得的电子探针的步进/束扫描进行分析，这特别适合于微量元素分析。长期以来，地球科学界一直需要石榴石中3+和4+阳离子的压力依赖性实验数据，因为涉及这些阳离子的地质过程发生在高压下。即将获得的实验数据将填补这一重大空白。由于Y3+的扩散率似乎比主要二价阳离子（Fe, Mg, Mn, Ca）的扩散率要慢得多，目前已有可靠的数据（Ca除外），因此，使用本研究过程中开发的计算机程序同时模拟石榴石中Y3+和二价阳离子的浓度分布，将对广泛的变质和岩浆过程的时间尺度施加严格的限制。变质岩中的石榴石已被广泛地用Sm-Nd和Lu-Hf年代学系统测定年代，但通常对同一块岩石的测定结果是不一致的。与Sm-Nd体系不同，母核素Lu3+和子核素Hf4+的扩散率似乎有很大差异。我们的初步实验数据似乎表明，这种性质和相对慢得多的Hf4+扩散率对石榴石全岩的Lu-Hf等时线的发展有有趣的影响，需要仔细考虑，并将在拟议的研究中加以解决，以正确解释年龄和176Lu/177Hf初始比，这是由自然样品的等时线得出的。此外，分析变质岩中石榴石的Sm-Nd和Lu-Hf年龄差异的扩散动力学性质有助于约束天然样品中石榴石的生长速率，这是模拟构造变质过程的一个重要参数。

Diffusion is a fundamental atomic process that underlies a variety of behavior of solids at relatively high temperatures, and thus is a subject of major interest in a number of fields in science and engineering. Although late in coming, there is now a major interest in the study of diffusion properties of rock-forming minerals as the modeling of diffusion-controlled properties have provided important constraints on the time scales of evolution of mountain belts, rate of subduction of tectonic plates, dynamical processes within the Earth's interior (e.g. melting relations and magma generation, convection), residence time scales of melts in magma chambers prior to eruptions that hold the potential of predicting volcanic activities etc. The proposed research is focused (a) on the experimental determination of diffusion kinetic properties of selected cations, namely Sm3+, Lu3+, Y3+ and Hf4+, in natural garnets as function of the important physical and chemical factors that influence diffusion process, and (b) applications of the data to the quantitative modeling of the time scales of a large spectrum of geological processes. Integration of these data with the results of earlier studies by the team on the diffusion properties of divalent (2+) cations in garnet would constitute a major step towards developing predictive models of diffusion in solids that is of fundamental importance to the general field of solid state diffusion. Depending on the sample and other details, the experimentally induced diffusion profiles would be analyzed by SIMS (secondary ion mass spectrometry) depth profiling in an Ion-probe, step scanning in a nanoSIMS or step/beam-scanning in our recently acquired electron microprobe, which is especially suited for trace element analysis. There has been a long standing need in the Earth sciences community for experimental data on the pressure dependence of 3+ and 4+ cations in garnets as the geological processes of interest involving these cations take place at high pressures. The experimental data to be obtained would fill this major gap. As the diffusivity of Y3+ seems to be significantly slower than those of the major divalent cations (Fe, Mg, Mn, Ca), for which reliable data are now available (except for Ca), simultaneous modeling of the concentration profiles of Y3+ and the divalent cations in garnet, using computer programs to be developed in the course of this study, would impose tight constraints on the time scales of a wide range of metamorphic and magmatic processes. Garnets in metamorphic rocks have been extensively dated by Sm-Nd and Lu-Hf geochronological systems, usually with discrepant results for the same rock. Unlike the Sm-Nd system, there seems to be a large difference between the diffusivities of parent nuclide, Lu3+, and the daughter nuclide, Hf4+. This property and the relatively much slower diffusivity of Hf4+, as our preliminary experimental data seem to suggest, have interesting implications on the development of garnet-whole rock Lu-Hf isochrons that need to be considered carefully, and would be addressed in the proposed research, for the proper interpretation of ages and 176Lu/177Hf initial ratios, which are derived from the isochrons in natural samples. Additionally, analysis of the discrepant Sm-Nd and Lu-Hf ages of garnets in metamorphic rocks in terms of the diffusion kinetic properties of the species would help constrain the growth rates of garnets in natural samples, which is a very important parameter for modeling tectono-metamorphic processes.