Experimental constraints on Li isotope fractionation during clay formation
Experimental constraints on Li isotope fractionation during clay formation
复制标题
DOI：
10.1016/j.gca.2019.02.015
复制
发表时间：
2019-04
影响因子：
5
通讯作者：
R. Hindshaw;Rebecca Tosca;Thomas L. Goût;I. Farnan;N. Tosca;E. Tipper
中科院分区：
文献类型：
--
作者：
R. Hindshaw;Rebecca Tosca;Thomas L. Goût;I. Farnan;N. Tosca;E. Tipper
摘要粘土矿物形成过程中锂同位素分馏因子是地球系统模型的一个关键参数。这项研究完善了我们对粘土形成过程中同位素分馏的理解，对野外数据的解释和Li的全球地球化学循环具有重要意义。我们合成富镁层状硅酸盐（硅镁石和皂石）在地球表面过程相关的温度。所得固体通过X射线衍射（XRD）和傅里叶变换红外光谱（FT-IR）表征以确认产物的矿物学和结晶度。散装固体样品用氯化铵处理，以除去可交换的锂，以区分这些网站和结构（八面体）网站之间的锂同位素分馏。与初始溶液相比，散装固体、残余固体和可交换溶液均富集6 Li。交换性溶液的δ 7 Li值平均比初始溶液低7‰。在20° C时，合成层状硅酸盐的残余固体和初始溶液δ 7 Li值（Δ 7 Li残余溶液）之间的平均差异为− 16.6±1.7‰，与模拟研究、高温实验数据和现场观察的外推结果一致。三个键合环境被确定从7 Li-NMR光谱，这是存在于本体和残余固体7 Li-NMR光谱，这意味着一些可交换的锂与氯化铵处理后仍然存在。7个Li-NMR峰归属于八面体、外层（层间和吸附）和伪六方（双三方腔）Li。通过结合7 Li-NMR数据与质量平衡约束，我们计算了分馏因子，基于蒙特卡罗最小失配方法，为每个键合环境。八面体、外球面和假六方晶位的计算值分别为− 21.5±1.1‰、− 0.2±1.9‰和15.0±12.3‰（误差1 σ）。体积分馏因子（Δ 7 Li本体溶液）取决于初始溶液的化学性质。初始溶液中Na浓度越高，体相δ 7 Li值越低。我们认为这是由于Na在层间位置的竞争中胜过Li，并且由于层间Li相对于八面体Li具有高的δ 7 Li值，因此增加的Na用于降低本体δ 7 Li值。在较高pH下进行的三个实验在残余固体中表现出较低的δ 7 Li值。这可能是动力学效应，由高pH下较高的反应速率引起，或者是平衡效应，由残余固体中Li掺入减少和/或溶液中Li形态变化引起。这项研究突出了7锂-核磁共振在粘土合成的实验研究的权力，以目标网站特定的锂同位素分馏因素，然后可以用来提供急需的现场过程的限制。
Abstract Knowledge of the lithium (Li) isotope fractionation factor during clay mineral formation is a key parameter for Earth system models. This study refines our understanding of isotope fractionation during clay formation with essential implications for the interpretation of field data and the global geochemical cycle of Li. We synthesised Mg-rich layer silicates (stevensite and saponite) at temperatures relevant for Earth surface processes. The resultant solids were characterised by X-ray diffraction (XRD) and Fourier-transform infrared spectroscopy (FT-IR) to confirm the mineralogy and crystallinity of the product. Bulk solid samples were treated with ammonium chloride to remove exchangeable Li in order to distinguish the Li isotopic fractionation between these sites and structural (octahedral) sites. Bulk solids, residual solids and exchangeable solutions were all enriched in 6 Li compared to the initial solution. On average, the exchangeable solutions had δ 7 Li values 7‰ lower than the initial solution. The average difference between the residual solid and initial solution δ 7 Li values (Δ 7 Li residue-solution) for the synthesised layer silicates was− 16.6±1.7‰ at 20° C, in agreement with modelling studies, extrapolations from high temperature experimental data and field observations. Three bonding environments were identified from 7 Li-NMR spectra which were present in both bulk and residual solid 7 Li-NMR spectra, implying that some exchangeable Li remains after treatment with ammonium chloride. The 7 Li-NMR peaks were assigned to octahedral, outer-sphere (interlayer and adsorbed) and pseudo-hexagonal (ditrigonal cavity) Li. By combining the 7 Li-NMR data with mass balance constraints we calculated a fractionation factor, based on a Monte Carlo minimum misfit method, for each bonding environment. The calculated values are− 21.5±1.1‰,− 0.2±1.9‰ and 15.0±12.3‰ for octahedral, outer-sphere and pseudo-hexagonal sites respectively (errors 1 σ). The bulk fractionation factor (Δ 7 Li bulk-solution) is dependent on the chemistry of the initial solution. The higher the Na concentration in the initial solution the lower the bulk δ 7 Li value. We suggest this is due to Na outcompeting Li for interlayer sites and as interlayer Li has a high δ 7 Li value relative to octahedral Li, increased Na serves to lower the bulk δ 7 Li value. Three experiments conducted at higher pH exhibited lower δ 7 Li values in the residual solid. This could either be a kinetic effect, resulting from the higher reaction rate at high pH, or an equilibrium effect resulting from reduced Li incorporation in the residual solid and/or a change in Li speciation in solution. This study highlights the power of 7 Li-NMR in experimental studies of clay synthesis to target site specific Li isotope fractionation factors which can then be used to provide much needed constraints on field processes.