Session chair: Tracey Jacksier, Air Liquide (US)
The session “Measuring isotopes” will take place on Wednesday 19 June 2019 with the following lectures:
E.01 – Enhanced mixture stability of stable isotopic gases in non-refillable cylinders by Tracey Jacksier, Air Liquide (US)
The dependence of stable isotopic analyses in geochemistry and environmental measurements are steadily increasing. However, the ability to compare measurements between laboratories can be quite challenging owing to differences in measurement and calibration. The key to obtaining reliable data is by designing experiments which utilize sampling methodologies that represent the environment intended for the study and meet the data-quality objectives. These samples in turn must then be calibrated against suitable reference materials containing low levels of uncertainty. The precision and accuracy of the analytical result is directly related to the precision and accuracy of the standard used to calibrate the analytical instrument. Isotopic primary reference materials, which utilize various laboratory extraction techniques (depending on the isotope of interest), are available in extremely limited quantities from the International Atomic Energy Agency (IAEA) and NIST. However, quantities of these materials are typically limited to one reference material per laboratory every 3 years. Laboratories are encouraged to develop their own working standards that can be used to calibrate samples. Preparation of these standards often increases the total uncertainty of the measurements. It is good analytical practice to matrix match the standard and sample as well as bracket the standards around the concentration of the samples. Ideally, 3 standards are the minimum required to assess linearity and accuracy. How can this be accomplished with the limited range of standards available? This presentation will examine the isotopic mixture preparation process in non-refillable cylinders, for both molecular and isotopic concentrations, for a range of components and delta values. The role of precisely characterized source material will be presented. Analysis of individual cylinders within multiple batches will be presented to demonstrate the ability to dynamically fill multiple cylinders containing identical compositions without isotopic fractionation. Additional emphasis will focus on the ability to adjust isotope ratios to more closely bracket sample types without the reliance on combusting naturally occurring materials, thereby improving analytical accuracy.
E.02 – Gas mixtures with stable 13C isotopes at different concentration by Fabrizio Turra, SIAD (IT)
The detection of carbon and oxygen isotopes are entering more and more in the analytical methods to identify the origin of food, to study the plant growth, to develop the petroleum geochemistry and for comprehensive understanding of pollution sources. The δ13C (‰ vPDB) ratio is usually in range from +5 to – 50 or lower. To have a complete set of stable CO2 isotopes concentration is important for the instruments calibration. Following the request from the users, a study to prepare mixture with δ13C of -100, using CO2 enriched by 12C, is under development. Once you have different carbon dioxide with the two isotopes concentration at the extremes, it is possible to prepare a carbon dioxide with any δ13C ratio in this range. The sequence of gas mixtures with δ13C at different ratio are prepared, starting from two carbon dioxide, using the same techniques adopted for the preparation of the metrological calibration mixtures to add low uncertainty to the mixtures. The usual preparation uncertainty is in the range of 0,3 % – 0,5 %: the final uncertainty, is thus, at the same level plus the initial measurement uncertainty of the two CO2 isotopes concentration. With the same procedures used in metrological field you can prepare not only the pure carbon dioxide at known isotopes concentration, but also other carbon compounds or prepare mixtures of these components in others gas balance at any concentration from 100 nmol/mol. Finally, it is important to have a stable isotopes concentration during the years and, thus, the raw material must be retained for long time to be sure to produce mixtures with the same δ13C. Up to now, in fact, the value of δ13C made from different laboratories is slightly different and to use exactly the same starting material gives confidence to a reliable δ13C value without variation over time.
E.03 – CO2 isotope ratio metrology: Measurement compatibility across IRMS inlet techniques by Abneesh Srivastava, NIST (US)
Commercially available isotope ratio mass spectrometry (IRMS) interface techniques allow for direct value assignment on the isotope Vienna Pee Dee Belemnite (VPDB) delta scale, across several sample form factors: pure CO2 via the dual inlet (DI) and CO¬2-air mixtures via the continuous flow (CF) method. CO2’s of common origin, as in pure CO2 and their derived CO2-in purified/synthetic air (serving as the dilution gas) mixtures, are expected to have identical isotope delta values. However, the process of measurement via varying sample-specific inlet techniques raises questions on the value agreement across techniques. At a broader level, due to the emergence of optical isotope ratio analyzers (to add to the existing option of mass spectrometry-based methods) comparability of measurements across techniques, sample form factor (pure CO2 versus derived CO2-air mixtures in pressurized cylinder), sample preparation method, sampling method comes into scrutiny. With the purported goal of cataloguing and deciphering the underlying mechanisms we are, in this study, conducting measurement comparability studies on pure CO2 using DI-IRMS and their derived CO2-in purified air (prepared at natural abundance concentrations) using CF-IRMS. Our presentation highlights: 1) Overview of CO2 gas isotope ratio measurement efforts in our group; 2) Establishment of a DI-IRMS method for δ13C, δ18O VPDB value assignment of pure CO2 as well as CF-IRMS method for the δ13C VPDB assignment of natural abundance CO2 in air cylinder mixtures; 3) δ13C VPDB value comparison across chosen IRMS inlet techniques.
E.04 – Preparation of pure CO2 samples for a CCQM comparison of isotope ratio measurement capabilities by Joële Viallon, BIPM (FR)
The characteristics and performance of an optimized calibration system for measurements of pure CO2 gas with isotope ratio infrared spectroscopy measurement systems that has allowed reproducible measurement of δ13C and δ18O values with 0,02 ‰ reproducibility (1 σ) is described. The key elements of the calibration system revolve around identical treatment of sample and reference gases allowing two-point calibration of up to 16 samples, and appropriate flushing protocols to remove any biases from memory effects of previously sampled gases. Measurements are performed by the IRIS system at a mole fraction of nominally 700 μmol/mol CO2 in air, by dilution of pure CO2 gas controlled by individual low-flow mass flow controllers (0,07 ml/min), and with a feedback loop to control mole fractions to ensure that differences between references and sample gas mole faction stay below 2 μmol/mol. This level of control is necessary to prevent biases in measured isotope ratios, the magnitude of which has also been studied with a sensitivity study that is reported as well. The system has been validated using pure CO2 samples which range in delta values of -1 ‰ and -45 ‰ vs VPDB, and in all cases measurement reproducibility over several days of testing of 0,02‰ or better (1 σ) were achieved for both δ13C and δ18O, with negligible memory effects. The amount of sample gas used for each measurement was less than 5 ml of CO2 at (RTP), making the system easily deployable for isotope ratio value assignment of bulk CO2 gas, and adaptable to atmospheric concentrations of CO2 in air, and for value assignments of standards. Using the calibration system described the measurement reproducibility of current IRIS systems is very close to the measurement performance that can be achieved with IRMS systems for δ13C measurements and potentially better than can be achieved for δ18O.