Simultaneous Analysis of Nitrogen, Carbon and Sulfur Stable Isotopes and Concentrations in Organics and Soils

Abstract:

To date, analysis of diet, food web complexities, biogeochemical cycles, and ecosystem functioning have largely focused on using variation in carbon (C) and nitrogen (N) stable isotope ratios. This is because a great deal is understood about what leads to this variation and because the dual stable isotope analysis of these two elements using continuous flow isotope ratio mass spectrometry (IRMS) is now commonplace. However, the aforementioned studies may all greatly benefit from the additional information one can get from also having sulfur (S) stable isotopes ratio data. Until very recently the analysis of δ³⁴S has traditionally required an additional and often more difficult analytical procedure.

Here, we report on the development of a new method that simultaneously analyzes the elemental and isotopic composition of N, C and S in a single sample. The new commercially available instrument includes a modified NCS elemental analyzer in line with an IRMS outfitted with 100 volt AD converters for wide dynamic range. We tested, and modified, this instrument to achieve maximum accuracy and precision for the isotopic measurements of all three elements. We found that the original design needed improvements to achieve our goals by: a) including a component (originally designed for trapping water) as buffer to reduce S memory and obtain reliable δ³⁴S analysis; b) adding an external furnace for complete reduction of nitrogen oxides to N₂ gas for accurate δ¹⁵N; c) adding a magnesium perchlorate water trap immediately after the reduction tube to minimize any water condensation that could also influence S memory. We analyzed a selection of organic materials and soils with approximately a 1:2 standards versus unknowns ratio per run.

Using this NCS set-up, the precision of the N and C isotopic measurements was comparable to the one usually attained in NC mode alone (standard deviation of ± 0.13 δ¹⁵N in the range 30 to 400 µg N, and of ± 0.12 δ¹³C in the range 0.20 to 4 mg C). The precision for S in the range 8 to 120 µg S was ± 0.4 δ³⁴S, although it decreased for samples with lower S content (± 1 δ³⁴S at 4 µg S).

Based on our testing, we believe that this method is suitable for widespread use, and can significantly enhance the application of δ³⁴S measurements in both soils and a broad range of organic samples in ecological and biogeochemical research.