This paragraph briefly summarizes standard laboratory procedures. Sample selection rigidly adheres to the following criteria to ensure top data quality – only euhedral grains, free of inclusions, and >70 µm in width (minimizing a-ejection correction). In our experience, the biggest hurdles to accurate and reproducible (U-Th)/He age determinations is the analysis of grains that are absolutely free of U- and Th-bearing inclusions. After careful microscopic grain selection/inspection (at 180x magnification), individual aliquots (3/sample) are photographed and measured for standard morphometric a-ejection age corrections. Subsequently, aliquots are wrapped in Pt foil tubes, laser heated for 5 minutes at 1070°C and analyzed for He, and reheated to ensure complete degassing. Zircon and all other silicate, oxide, and phosphate mineral phases are heated for 10 minutes at 1300°C and subsequently reheated until completely degassed (<1% He re-extract). After laser degassing, radiogenic He is analyzed in an all-metal, fully computer-automated UHV He extraction line that is equipped with precise volume aliquot systems for 3He isotope dilution and delivering of 4He standard gas, a cryogenic gas purification system capable of separating He from other gases by variable temperature release at 16-37K, and a Blazers Prisma QMS-200 quadrupole mass spectrometer for measuring 3He/4He ratios. Our helium isotope dilution procedure allows very low-blank and high-precision He measurements (~0.3-0.5%). The primary KU laser He extraction line is capable of analyzing up to 36 samples in a fully automated fashion. He gas standards and He extraction line blanks are routinely analyzed as part of unknown age analyses.
After complete degassing, aliquots are retrieved and dissolved for U, Th, and Sm determination. Dissolution procedures vary greatly and can be very time consuming (e.g., zircon or rutile). In the case of apatite, samples are spiked (230Th, 235U, 149Sm) and dissolved in 30% HNO3 (90°C for 1 hour). Zircon and rutile samples are unwrapped from Pt foil (PtAr inference with U) and dissolved using standard U-Pb double pressure-vessel digestion procedures (HF- HNO3 and HCl) for a total of 4 days. Titanite, monazite, and magnetite are dissolved in a HCl-HF mixture (150°C for 24 h). Rutile and magnetite undergo ion-exchange column chemistry procedures to eliminate interfering reactions with Ti and Fe. After dissolution, spiked aliquot solutions are analyzed for U, Th, and Sm using the Thermo Element2 HR-ICP-MS, fitted with a CETAC micro-concentric nebulizer and ESI autosampler.
Data reduction has been streamlined in light of the tremendous sample throughput (>7000 analyses in 2010) with all data (morphometric analyses to He and U, Th, and Sm analyses) compiled and reduced using in-house designed user-friendly visual-basic ExcelTM add-in macro software packages. All data and metadata are centrally stored in a laboratory database.
All analytical uncertainties are captured and properly propagated during multi-step and multi-instrument (U-Th-[Sm])/He analysis. Analytical uncertainties include easily quantifiable variables such as He measurement error (0.3-0.5%) and U, Th, and Sm measurement error (<1-2%). More difficult to assess are uncertainties related to a-ejection corrections due to inherit assumptions about U and Th distribution. As a result propagated analytical uncertainties (~3-4%, 2-s) generally underestimate aliquot reproducibility. Therefore, a commonly employed approach in (U-Th)/He dating is the assignment of a percentage error to an aliquot analysis based on the standard deviation of a population derived from standard samples; for example 6% for apatite and 8% for zircon. For the proposed work both propagated errors and reproducibility-based standard deviation errors will be reported.