Methods for the microsampling and high-precision analysis of strontium and rubidium isotopes at single crystal scale for petrological and geochronological applications

Micromilling of single crystals to yield microgram-sized solid samples for subsequent isotopic analysis can yield important petrogenetic information from crystals (especially feldspars) in magmatic rocks. Exceptional detail can be gained because the textural context of the sample area can be fully evaluated before milling. Such information is otherwise unknown or lost when performing analysis at the bulk rock scale. Here we present a comprehensive methodology for the precise analysis of ng-levels of Rb and Sr purified from microgram-sized solid samples extracted from minerals in thin sections. Physical sampling techniques are based around a computer numerical control (CNC) milling machine (Micromill™), newly designed specifically for the sampling of complex accretionary and growth structures in crystalline materials. Chemical procedures for the isolation of Sr and Rb for TIMS and MC-ICPMS analysis, respectively, are presented, and the performance of these analytical techniques is assessed in the context of the small sample sizes produced by micromilling. Despite being time consuming, mechanical sampling, conventional dissolution and chemical separation followed by analysis by TIMS remain the best methods for both highly accurate and precise determinations of Sr isotopic compositions in most geological materials over a wide range of Sr concentrations, Rb/Sr ratios and matrix types. Using these techniques, it is possible to achieve long-term 2 S.D. external precisions of 50 ppm for load sizes as small as 3 ng Sr. We demonstrate the validity of these techniques using two examples. The first is from a < 50 ka single feldspar crystal from Parinacota volcano (Chile) to show that 87Sr/86Sr isotope ranges as small as 0.00006 (2 × 2 S.D. external) can be resolved under conditions where radiogenic-Sr ingrowth can be neglected. The second is from the 28.4 Ma Fish Canyon Tuff (Colorado) to demonstrate the utility of isotope dilution measurements of Rb and Sr contents to calculate 87Rb/86Sr and thus perform age corrections to the 87Sr/86Sr ratios in order to establish 87Sr/86Sri variations between single crystals or zones. We demonstrate that biotite crystals in the Fish Canyon Tuff show Sr-isotopic variations well beyond analytical errors and hence that the crystals involved were not in isotopic equilibrium and cannot be used to establish an isochron age. On the other hand, the precision of our isotope dilution measurements may be used to obtain Rb–Sr geochronological information at the single crystal scale, provided that homogeneity in 87Sr/86Sr at the time of crystallisation can be demonstrated.