| Ca isotopic composition is an important tracer in chemical weathering, mantle evolution, and biochemical processes. The variation of Ca isotopic composition (expressed as δ44Ca) is about 4%o in terrestrial samples, but changes of Ca isotopic compositions in natural rocks are relatively small. Therefore, highly precise and accurate methodologies are needed to investigate these small isotopic differences in natural samples.Two analytical approaches have been frequently used for Ca isotopes analysis: double-spike method using thermal ionization mass spectrometry (DS-TIMS) and standard-sample-standard bracketing method using multi-collector inductively coupled plasma mass spectrometry (SSB-MC-ICP-MS).The DS-TIMS method has an advantage of higher precision compare to the SSB-MC-ICP-MS method, since optical interferences are smaller and double-charge and molecular interferences are negligible in TIMS. Moreover, the mass fractionation during chemical separation can be corrected by DS-TIMS method. As a result, Ca stable isotopic compositions are most frequently measured by DS-TIMS.In this study, we focused on the optimization of the Ca chemical separation and the TIMS measurement conditions aiming to obtain high precision Ca isotope analysis of geologic samples, and the main achievements are summarized as follows:(1) A novel Ca purification method is established. As opposed to common cation exchange resin, a micro-column filled with Ca selective resin (DGA resin) is used to achieve high recovery and efficient separation of Ca from the sample matrix (Ti, K, Al, Fe, Mg). The matrix and analyte are concentrated in two small fractions,1.4 ml 3 N HN03 and 5ml DI-water, respectively. Furthermore, three distinct sample matrices (synthesis, BCR-2 and seawater) are tested to evaluate the separation efficiency and the Ca recovery with use of DGA column, and the resluts comfirmed the efficiency of our Ca purification method for different matrix. (2) We present an efficient model for theoretically predicting the precision of double spike technique when peak jump mode is employed with use of Monte Carlo simulation technique. Previous studies have focused on the optimization of the double spike composition and the spike proportion in sample mixture with use of static collector model. However, our results obtained in this study show that the measurement precision of δ44Ca ration can be further reduced by use of optimum ratio combination and cup configuration. Use of 42Ca-48Ca double spike (proportion of 42Ca in the double spike is 44% while proportion of double spike in the mixture is 12%) and 40Ca#1/44Ca#1,42Ca#1/44Ca#1,48Ca#2/44Ca#2 cup configuration with 4 s integration time in the first sequence and 12 s in the second sequence is recommended to achieve the best measurement precision of δ44Ca ratio by DS-TIMS. With using these optimal parameters, predicted precision is improved by 25% compared to that estimated under static model. The observed measurement precision of δ44Ca in standard samples are in good agreement with the predicted precision, validating the effectiveness of the Monte Carlo model for error estimation. Our method is expected to be applicable to other isotopes which cannot be measured simultaneously in a single line by DS-TIMS.Using the optimal 42Ca-48Ca DS-TIMS technique, Ca isotopic compositions of sixteen Ca-bearing USGS geological reference materials with different matrices are determined in two laboratories (GPMR:State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan and CIG:Center for Isotope Geochemistry, University of California, Berkeley). The long-term reproducibility is evaluated at 0.15%o (2σ) for δ44Ca. at both laboratories, based on replicate measurements of pure Ca isotope reference material NIST SRM 915a, NIST SRM 915b and seawater. Results in this study contribute to a better characterization of Ca isotopic compositions of different sample matrices. Moreover, high quality Ca isotope data for one carbonatite reference materials (COQ-1) and three sedimentary reference materials (SBC-1, NOD-P-1, and NOD-A-1) are reported for the first time. Where comparison is available, the Ca isotopic compositions of USGS reference materials are not only in agreement between GPMR and CIG but also in agreement with previously published data within quoted uncertainties. The comprehensive dataset reported in this study serves as a reference for both quality assurance and inter-laboratory comparison of high precision Ca isotopic study. Meanwhile, the data presented in this study show that significant variability is present in Ca isotopic compositions of igneous rocks, suggesting that Ca isotopes fractionate during high temperature geological processes, with ultramafic rocks having heavier Ca isotopic compositions (1.20‰-1.51‰) and mafic to felsic rocks having lighter Ca isotopic compositons (0.70‰-0.89‰). |
PDF Full Text Request