Dear Editor, Since the pioneering work by Emiliani and Shackleton, the carbon and oxygen stable isotopes of carbonates have been of major importance in geological and environmental sciences.[1-8] Palaeoceanographic and terrestrial climate research (e.g. on secondary cave carbonates or lacustrine sediments) routinely make use of the δ13C and δ18O of carbonates as climate proxies.[9-12] Online sample preparation and continuous-flow isotope ratio mass spectrometry (IRMS) with multi-loop injection procedures are now widely applied analytical techniques, because they allow precise and accurate measurements at high sample throughput.[13] The samples sizes used in analytical systems for carbonates described in the literature are generally larger than ca. 50 µg, with an external precision of ca. 0.05–0.10‰.[14, 15] Unfortunately, sample size is often a limiting factor in studies of many climate records (lacustrine sediments, speleothems, foraminifera) when sampling for individual sediment layers to obtain sub-annual to seasonal information.[16, 17] Reliable, accurate, and fast analysis of samples as small as possible is desired for ultra-high-resolution sampling for palaeoclimate investigations, but also for other geological applications, such as investigations of traces of carbonates in silicate rocks.[9, 18-20] Fiebig and co-workers described a method for the precise measurement of carbonate samples in the range 10 to 30 µg for the GasBench II, the same device as we describe in this communication.[21] However, their method requires a modification of the hardware and the use of liquid nitrogen, which increases the costs and the complexity of the measurements. In this letter we describe the improvements in analytical performance possible by simply reducing the size of sample vials from 12 to 4.5 mL and show that precise and accurate measurements on samples as small as 20 µg of carbonate are possible. For the experiments described below we employ the ThermoFinnigan (Bremen, Germany, now Thermo Fisher Scientific) GasBench II, equipped with a CTC autosampler (CTC Analytics AG, Zwingen, Switzerland), and coupled to a ConFlow IV interface and a Delta V Plus mass spectrometer (both Thermo Fisher Scientific) at the ETH Zurich (Zurich, Switzerland). The system setup is essentially the same as used in an earlier study.[13] Instead of the 12 mL exetainers normally used, we load micro-samples (7–150 µg) of carbonate into 4.5 mL round-bottomed borosilicate vials (Labco, High Wycombe, UK, part No. 948 W), capped with Labco butyl rubber septa.[13] Samples were weighed using a Mettler Toledo MT5 FACT microbalance (reproducibility 0.8 µg; Leicester, UK). The exetainers were flushed for 240 s on a model 222XL autosampler (Gilson™, Middleton, WI, USA), with helium grade 5.0 (99.999% He) at a flow rate of 55 mL/min, corresponding to a flushing volume of 220 mL. The smaller 4.5 mL exetainers allow the reduction of helium consumption compared with the 12 mL vials, as no air contamination was observed during our experiments (Fig. 1). Eight-one exetainers (including 18 in-house standards (powdered Carrara marble, MS2) are subsequently heated to 72 ± 0.1 °C in the aluminium block of the autosampler and acid-digested following standard procedures.[13] The heating period of the autosampler heating block is somewhat longer for short vials, due to the air in the heating block below the vials. If small vial runs are planned for an extended period of time, metal cylinders placed below the vials can be used to circumvent the problem.
Search from 250 M+ research papersSearch from 250 M+ research papers
Feb 3, 2012
Martin Elsner + 6
Open Access