TILDAS measurement of multiple clumped isotope ratios in carbonates: progress and new horizons

Isotopic analysis using high resolution Isotope Ratio Laser Spectroscopy (IRLS) has been shown to be advantageous to multiple geochemical applications during the last decade.  These advances include isotopic analysis of the bulk isotopic compositions of water, carbon dioxide, methane and nitrous oxide.  More recently, laser spectroscopy has been used by several groups to examine the isotopic compositions of methane, carbon dioxide and nitrous oxide when carrying two rare isotopes (so called clumped isotopes).   Our recent work using Tunable Infrared Laser Direct Absorption Spectroscopy (TILDAS) has demonstrated highly accurate (~0.01 ‰) measurements of the clumped (¹⁶O¹³C¹⁸O or 638 in HITRAN isotope notation) isotopic composition of carbon dioxide derived from carbonate samples with spectroscopic measurement times of ~30 minutes using a dual laser spectrometer.  That spectrometer is optimized for the measurement of the four isotopologues required to calculate Δ₆₃₈ (or Δ₄₇).  We present here our parallel project to develop a novel dual laser isotope analyzer capable of measuring multiple carbon dioxide isotopic signatures simultaneously.   Specifically, we simultaneously measure the isotopic abundances of the three most abundant clumped isotopologues (Δ₆₃₈, Δ₆₃₇ and Δ₈₂₈) as well as ¹⁷O oxygen excess or Δ¹⁷O.  Δ₆₃₈ and Δ₈₂₈ correspond to the quantities Δ₄₇ and Δ₄₈ when measured by isotope ratio mass spectroscopy (IRMS).   Δ¹⁷O is very difficult to measure with IRMS and Δ₆₃₇ has not been previously measured with any technique to the best of our knowledge.  The new instrument utilizes carefully chosen spectral windows, operates at low sample pressure and exploits automated laser frequency hopping.  This prototype instrument simultaneously measures seven isotopologues of carbon dioxide: 626, 636, 628, 627, 638, 637 and 828.  Our preliminary results for Δ₈₂₈ (or Δ₄₈) are displayed as an Allan-Werle plot which shows that the precision in the measurement of Δ₈₂₈ is ~0.09‰ for a single 3 minute sample measurement referenced to a working reference gas.  The plot shows that instrumental drift is very small over periods of several hours and that the precision can be improved to 0.03‰ by processing 10 sub-samples or to 0.01‰ by processing 100 sub-samples.  These measurements are preliminary and somewhat idealized but show promise for this new technique.