Discrimination of TiO2 polymorphs in sedimentary and metamorphic rocks
Discrimination of TiO2 polymorphs in sedimentary and metamorphic rocks
复制标题
DOI：
10.1007/s00410-010-0551-x
复制
发表时间：
2011-04
影响因子：
3.5
通讯作者：
Silke Triebold;G. Luvizotto;R. Tolosana-Delgado;T. Zack;H. Eynatten
中科院分区：
文献类型：
--
作者：
Silke Triebold;G. Luvizotto;R. Tolosana-Delgado;T. Zack;H. Eynatten
通过对不同地质背景下的样品进行拉曼光谱研究发现，除了金红石之外，几乎无法预测TiO 2多晶型物的存在，此外，金红石的存在比以前认为的更普遍。变质作用的压力和温度以及全岩化学控制了金红石的产出，而矿物组合特征和/或流体产状或成分的变化则影响了金红石微量元素特征和残余金红石的再平衡。微量元素含量的评估，通过电子探针在钛铁矿，板钛矿，金红石显示，这些变化显着的三个TiO 2相之间。因此，一方面，如果不事先进行物相鉴定，仅根据Nb和Cr含量来确定烃源岩类型，以及根据Zr含量来进行测温，都会导致错误的结果。对于元素Cr、V、Fe和Nb，多晶型物之间的变化是系统的，并且可以用于基于线性判别分析的区分。使用相组的手段和系数的线性判别式从汇编的分析与定义明确的相位信息的样品连同先验概率的分组从一个自然的样品汇编，一个是能够计算的任何TiO 2分析包含至少的关键元素Cr，V，Fe和Nb的相位分组概率。应用结果表明，对金红石相的归属，分类正确率可达99.5%。因此，通过痕量元素的相位规格被证明是除了拉曼光谱之外的有价值的工具。
Investigation by Raman spectroscopy of samples from different geological settings shows that the occurrence of TiO2polymorphs other than rutile can hardly be predicted, and furthermore, the occurrence of anatase is more widespread than previously thought. Metamorphic pressure and temperature, together with whole rock chemistry, control the occurrence of anatase, whereas variation of mineral assemblage characteristics and/or fluid occurrence or composition takes influence on anatase trace element characteristics and re-equilibration of relict rutiles. Evaluation of trace element contents obtained by electron microprobe in anatase, brookite, and rutile shows that these vary significantly between the three TiO2phases. Therefore, on the one hand, an appropriation to source rock type according to Nb and Cr contents, but as well application of thermometry on the basis of Zr contents, would lead to erroneous results if no phase specification is done beforehand. For the elements Cr, V, Fe, and Nb, variation between the polymorphs is systematic and can be used for discrimination on the basis of a linear discriminant analysis. Using phase group means and coefficients of linear discriminants obtained from a compilation of analyses from samples with well-defined phase information together with prior probabilities of groupings from a natural sample compilation, one is able to calculate phase grouping probabilities of any TiO2analysis containing at least the critical elements Cr, V, Fe, and Nb. An application of this calculation shows that for the appropriation to the phase rutile, a correct-classification rate of 99.5% is obtained. Hence, phase specification by trace elements proves to be a valuable tool besides Raman spectroscopy.