13 δ C Measuring the Signature of Dissolved Organic Carbon: An Inter-laboratory Comparison Study Andrew Barber, Karine Lalonde, Yves Gélinas GEOTOP and Concordia University, 7141 Sherbrooke West, Montréal, Quebec, H4B 1R6, Canada DOC Oxidation Methods Objectives We called upon a number of laboratories to participate in an inter-comparison study to evaluate the applicability of these methods in measuring: 1. [DOC] 2. δ13C of DOC For: 1. Standards (known δ13C) and samples (unknown δ13C) 2. Freshwater and saltwater solutions High temperature catalytic oxidation Sample CO2 Wet oxidation Sample CO2 LC-DOC Using these samples we can compare the ability of each oxidation method to measure the δ13C-DOC of simple solutions at different concentrations S2O82S2O82Sample Pt H3PO4 Reactor μHPLC • Up to 1.8 mL injections • Tens of mLs of sample • 100 - 200 μL injections High Temperature Catalytic Oxidation G. G. Hatch Lab (University of Ottawa) UC Davis Stable Isotope Facility Joy Matthews USGS / WHOI John Pohlman Southern Cross University KU Leuven FAU Erlangen-Nuremberg Wet Chemical Oxidation Vrije Universiteit Brussel NC State University WCO coupled to Gas Bench UC Davis Stable Isotope Facility ETH Zürich Université d’Orléans LC-IRMS NIOZ MARUM Cavity Ring-Down Spectrometer Solid Phase Extraction + EA-IRMS Andrew Barber /Karine Lalonde Paul Middlestead University of Malaga ICBM Oldenburg Acknowledgements Matheus Carvalho Steven Bouillon Robert Van Geldern CRDS (n = 1) KHP (-28.16‰) -30 HTC-DOC-IRMS WCO-IRMS LC-IRMS SPE-EA-IRMS 75 µM 393 µM -25 500 -26 400 -27 300 -28 200 -29 100 -30 Lake UDOM (-27.50‰) 0 0 200 400 600 800 True Concentrations (µmol L-1) Figure 1 : Reported versus true (Solid line) of [DOC] measurements depending on oxidation method used [DOC] for Saline Samples (µmol L-1) 42 Xavier Prieto /Marcus Elvert -18 58 201 246 557 -20 -22 -24 HTC 1 -26 -28 HTC 2 HTC 3 WCO 1 Figure 2 : δ13C of saline solutions reported by various users (3 HTC-DOC users and 1 WCO user). Dotted lines represent “true” value, when available www.PosterPresentations.com -29 -24 LC-DOC (n = 3) -30 RESEARCH POSTER PRESENTATION DESIGN © 2012 -28 -23 600 Marco Houtekamer We would like to express our sincerest gratitude to all participants and ISOGEOCHEM frequenters who used their precious instrument time to analyze the samples or personal time for discussing this project. -27 WCO (n = 4) 700 -16 Thorsten Dittmar/ Jutta Niggemann HTC-DOC (n = 4) 800 Patrick Albéric Iñaki Vadillo 74 µM 125 µM 559 µM 773 µM -22 900 Chris Osburn Susan Lang [DOC] for All Samples 1000 Natacha Brion Joy Matthews SPE-EA-IRMS -26 Analysts Measured Concentrations(µmol L-1) Concordia University LC-IRMS -24 • 680-720oC conversion • Persulfate chemically • Persulfate chemically of DOC in the solidoxidizes DOC (100oC) oxidizes DOC (100oC) state/gas phase • Competition between • Competition between • Salt/matrix does not DOC and Cl- for the DOC and Cl- for the affect combustion persulfate reagent persulfate reagent δ13C – DOC (‰) Laboratory/ Institution WCO-IRMS -25 List of participants Oxidation Method Used HTC-DOC-IRMS CO2 100oC 720oC δ13C Signature of KHP and Lake UDOM Samples (Freshwater) δ13C DOC (‰) Much progress has been made towards precisely and accurately measuring the concentration of dissolved organic carbon (DOC) in the past 4 decades. A number of oxidation methods have been developed, including: 1. High-temperature catalytic oxidation (HTC-DOC–IRMS) 2. Wet chemical oxidation (WCO-IRMS) 3. Liquid chromatography (LC-IRMS) 4. Solid phase extraction (SPE) followed by EA-IRMS A set of four Potassium Hydrogen Phthalate (KHP) solutions and 2 lake UDOM samples (known δ13C) with varying DOC concentrations were included in the sample set unbeknownst to the analysts. Figure 3 : Reported δ13C of KHP solutions and Lake UDOM samples as a function of oxidation method and solution concentration (μM). True δ13C values are shown as dotted lines Conclusions & Recommendations • [DOC] can be determined accurately for all methods but δ13C remains a difficult measurement, especially for low [DOC] solutions; • Salty solutions were not analyzed by most participants because of analytical limitations. Only 3 HTC instruments and 2 WCO instrument measured the δ13C of the total DOC pool; • Samples with mixed DOC sources (2 pools with very different δ13C signatures) generated the highest standard deviations, possibly due to incomplete combustion of one of the DOC pools; • Routine use of a complex natural sample with a known δ13C signature is recommended to verify instrument performance.