Calcium isotope measurements using a collision cell (CC)-MC-ICP-MS

Calcium is a major element in terrestrial planets and has six stable isotopes, including Ca-40 which is a product of radioactive decay of K-40 (T-1/2 = 1.25 Gyrs). Calcium isotope geochemistry has therefore been used to trace biological and geological processes in various environments as well as radiogenic dating. However, it has been technically impossible to analyze stable and radiogenic isotopes simultaneously and high precision measurements usually require tens of micrograms of Ca. Here, we present high-precision Ca isotopic data, including the stable isotopes Ca-42, Ca-43 and Ca-44 as well as the radiogenic Ca-40 using a Nu Sapphire multi-collector inductively-coupled-plasma mass-spectrometer (MC-ICP-MS), that is equipped with a collision cell to minimize Ar based interferences. The stable isotopic variations are reported using the delta notation (delta Ca-40/44, delta(42/44)ca, delta Ca-43/44), which is the permit deviation of a given ratio from the NIST SRM 915b standard. The radiogenic excess is reported as epsilon Ca-40, the per ten thousand deviation of the Ca-40/Ca-44 ratio from the same standard, after internal normalization to the Ca-42/Ca-44 ratio. We tested the influence of various parameters (Ca concentration, Ca and HNO3 concentration mismatch between sample and standard, effect of matrix elements) on the precision and accuracy of the measurements. We show that, using similar to 100 ng is sufficient to obtain high precision (<100 ppm, 25D) delta Ca-40/44 and epsilon Ca-40 using a standard bracketing technique-this represents an improvement by over an order of magnitude in sensitivity compared to any previous methods. It should be noted that a limitation is that the intensities of the standard and sample must be matched very closely (better than 2%). Following this recommendation, the Ca isotopic compositions of 9 geological samples reported here is consistent with previous data. For example, the epsilon Ca-40 value of GSP-2 (3.92 +/- 0.48 parts per thousand) is within error with literature data (values between 3.69 and 4.32 parts per thousand). Using this method, it is therefore possible, for the first time, to obtain simultaneously high-precision radiogenic and stable Ca isotopes on small samples, which will certainly open up many novel applications of Ca stable isotopes.