THE VISCOSITY OF PLANETARY THOLEIITIC MELTS: A CONFIGURATIONAL ENTROPY MODEL
THE VISCOSITY OF PLANETARY THOLEIITIC MELTS: A CONFIGURATIONAL ENTROPY MODEL
复制标题
行星拉斑石熔体的粘度：构型熵模型
DOI：
10.1130/abs/2016am-279885
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
A. Sehlke
中科院分区：
文献类型：
--
作者：
A. Whittington;A. Sehlke
硅酸盐熔体的粘度（η）是控制岩浆体系中物质传递的基本物理性质。粘度可以跨越许多数量级，强烈依赖于温度和组成。有几个模型可以很好地描述这种对陆地融化的依赖性。然而，行星玄武岩熔岩的成分明显不同，主要是贫碱，富铁和/或高镁。用同心圆筒和平行板粘度计测量了20个拉斑玄武岩熔体的粘度，其中15个代表火星、水星、月球、木卫一和灶神星已知或估计的表面成分。行星玄武岩在液相线温度下的粘度范围为2个数量级，在玻璃化转变附近的粘度范围为4个数量级，并且可以比陆地熔岩更粘或更粘。我们发现，目前的模型下，高估超液相线粘度高达2个数量级的这些组合物，偏离甚至更强烈地从测量的玻璃化转变粘度。我们用Adam-Gibbs理论（A-G）将粘度（η）与绝对温度（T）和该温度下体系的构型熵（Scoff）联系起来，其形式为log η = Ae + B e /TSCoff。我们研究的组合物的玻璃和液体的热容（CP）通过可用的文献模型计算。我们表明，A-G理论是适用于模拟个别复杂拉斑玄武岩熔体的粘度含有10个或更多的主要氧化物以及或更好地比常用的经验方程。我们使用-3.34 ± 0.22 log单位的常数Ae和12个可调子参数成功地模拟了全局粘度数据集，这些参数捕获了熔体粘度对组成和温度的依赖性。七个子参数占的成分依赖性的Be和5 Sconf。我们的模型再现了496个测得的粘度数据点，1σ均方根偏差（rmsd）为0.12 log单位，跨越13个测得的熔体粘度的订单。该模型在预测月球和火星熔融物的粘度方面表现良好，但没有用于校准，应该用于计算其他卫星和行星上无水玄武岩火山活动的熔岩流速度和通量。
Abstract The viscosity (η) of silicate melts is a fundamental physical property controlling mass transfer in magmatic systems. Viscosity can span many orders of magnitude, strongly depending on temperature and composition. Several models are available that describe this dependency for terrestrial melts quite well. Planetary basaltic lavas however are distinctly different in composition, being dominantly alkali-poor, iron-rich and/or highly magnesian. We measured the viscosity of 20 anhydrous tholeiitic melts, of which 15 represent known or estimated surface compositions of Mars, Mercury, the Moon, Io and Vesta, by concentric cylinder and parallel plate viscometry. The planetary basalts span a viscosity range of 2 orders of magnitude at liquidus temperatures and 4 orders of magnitude near the glass transition, and can be more or less viscous than terrestrial lavas. We find that current models under- and overestimate superliquidus viscosities by up to 2 orders of magnitude for these compositions, and deviate even more strongly from measured viscosities toward the glass transition. We used the Adam–Gibbs theory (A–G) to relate viscosity (η) to absolute temperature (T) and the configurational entropy of the system at that temperature (Sconf), which is in the form of log η = A e + B e / TS conf . Heat capacities (CP) for glasses and liquids of our investigated compositions were calculated via available literature models. We show that the A–G theory is applicable to model the viscosity of individual complex tholeiitic melts containing 10 or more major oxides as well or better than the commonly used empirical equations. We successfully modeled the global viscosity data set using a constant Ae of −3.34 ± 0.22 log units and 12 adjustable sub-parameters, which capture the compositional and temperature dependence on melt viscosity. Seven sub-parameters account for the compositional dependence of Be and 5 for Sconf. Our model reproduces the 496 measured viscosity data points with a 1σ root-mean-square deviation (rmsd) of 0.12 log units across 13 orders of measured melt viscosity. The model performed well in predicting the viscosity of lunar and martian melts not used in calibration, and should be used to calculate lava flow velocities and fluxes for anhydrous basaltic volcanism on other moons and planets.