Isotopically characterised N2O reference materials for use as community standards

Rationale Information on the isotopic composition of nitrous oxide (N2O) at natural abundance supports the identification of its source and sink processes. In recent years, a number of mass spectrometric and laser spectroscopic techniques have been developed and are increasingly used by the research community. Advances in this active research area, however, critically depend on the availability of suitable N2O isotope Reference Materials (RMs). Methods Within the project Metrology for Stable Isotope Reference Standards (SIRS), seven pure N2O isotope RMs have been developed and their 15N/14N, 18O/16O, 17O/16O ratios and 15N site preference (SP) have been analysed by specialised laboratories against isotope reference materials. A particular focus was on the 15N site‐specific isotopic composition, as this measurand is both highly diagnostic for source appointment and challenging to analyse and link to existing scales. Results The established N2O isotope RMs offer a wide spread in delta (δ) values: δ 15N: 0 to +104‰, δ 18O: +39 to +155‰, and δ 15NSP: −4 to +20‰. Conversion and uncertainty propagation of δ 15N and δ 18O to the Air‐N2 and VSMOW scales, respectively, provides robust estimates for δ 15N(N2O) and δ 18O(N2O), with overall uncertainties of about 0.05‰ and 0.15‰, respectively. For δ 15NSP, an offset of >1.5‰ compared with earlier calibration approaches was detected, which should be revisited in the future. Conclusions A set of seven N2O isotope RMs anchored to the international isotope‐ratio scales was developed that will promote the implementation of the recommended two‐point calibration approach. Particularly, the availability of δ 17O data for N2O RMs is expected to improve data quality/correction algorithms with respect to δ 15NSP and δ 15N analysis by mass spectrometry. We anticipate that the N2O isotope RMs will enhance compatibility between laboratories and accelerate research progress in this emerging field.

[Correction added on 16 May 2022, after first online publication: CSAL funding statement has been added.] Results: The established N 2 O isotope RMs offer a wide spread in delta (δ) values: Conclusions: A set of seven N 2 O isotope RMs anchored to the international isotoperatio scales was developed that will promote the implementation of the recommended two-point calibration approach. Particularly, the availability of δ 17 O data for N 2 O RMs is expected to improve data quality/correction algorithms with respect to δ 15 N SP and δ 15 N analysis by mass spectrometry. We anticipate that the N 2 O isotope RMs will enhance compatibility between laboratories and accelerate research progress in this emerging field.

| INTRODUCTION
Since its first application by Sakae Toyoda and Naohiro Yoshida in 1999, 1 site-specific N 2 O isotope analysis has been applied by many research groups to differentiate N 2 O source and sink processes at different spatio-temporal scales (see reviews by Toyoda et al, 2 Ostrom et al, 3 Decock et al, 4 Denk et al, 5 and Yu et al 6 ). Likewise, dual-isotope plots (e.g. δ 15 N SP /δ 15 N) or so-called "isotope mapping" approaches have been used to constrain the contributions of specific pathways, and the effect of isotope fractionation during N 2 O reduction. 7,8 The informative value of N 2 O isotope data has been markedly increased by using the data to inform biogeochemical models, providing regional and global patterns of N 2 O losses and independent process information. [9][10][11][12] Advances in applications have been accompanied and accelerated by progress in analytics, complementing the traditional high-precision isotope-ratio mass-spectrometry (IRMS) 1,13 by laser spectroscopic techniques, with the potential for field applicability and real-time data coverage. [14][15][16][17][18][19] The isotopic composition of a sample is reported using the delta (δ) notation, which is the relative difference in isotope ratio (R) between a sample P and a reference material, i.e. δ(P/ref) = R P / R ref À 1. For nitrogen, the 15  Further progress in N 2 O isotope research critically depends on the compatibility of laboratory results. 20 To achieve this, individual laboratories have to implement a traceability chain, i.e. a hierarchy of reference materials which descends with increasing uncertainty, linking the isotopic composition of primary RMs used to realise the respective scale, through secondary standards and working laboratory standards to a sample. 21 Generally, two RMs with distinct δ values should be used for calibration purposes, following the two-point data normalisation requirement. However, primary RMs and secondary scale anchors for δ 15 N (ammonium sulfate, potassium nitrate) as well as δ 17 O and δ 18 O (water) have a different chemical identity than N 2 O sample gas. Thus, a chemical conversion reaction 20 has to be implemented prior to analysis, which requires specialised laboratories.
The synthesis of N 2 O by thermal decomposition of isotopically characterised ammonium nitrate (NH 4 NO 3 ) has been suggested as an approach to link the position-dependent nitrogen isotopic composition of N 2 O to the Air-N 2 scale. 1 The basic concept of this technique is that the nitrogen atom at the α-position of of the formed N 2 O originates from NO 3 À , while the β-nitrogen comes from NH 4 + . 22 The validity of the NH 4 NO 3 decomposition technique has been confirmed, 23,24 but its accuracy for the calibration of δ 15 N α and δ 15 N β was found to be limited by non-quantitative NH 4

| EXPERIMENTAL
The main purpose of this study is the provision of isotopically characterised N 2 O RMs, covering an extended range of delta values as compared to existing gases. Figure 1 provides a schematic overview on the links established within this study between existing international RMs and the novel gaseous In section 2.1 ("left branch" of Figure 1), 15 T A B L E 2 Overview of NH 4 NO 3 salts (S1-S6) prepared from commercially available NH 4 NO 3 (A-E) and covering a wide range of δ 15 N(NH 4 + ) and of substrate NH 4 NO 3 salts (S1-S6) and were analysed together with S1-N 2 O and S4-N 2 O, as described in the preceding section, to propagate the moiety-specific isotopic composition defined by S1 and S4 to the novel RMs (Equation 1). An uncertainty assessment was conducted according to Equation 2 including uncertainties of S1-N 2 O and S4-N 2 O, as discussed above, their analyses, and the analyses of RMs.   The isotopic composition of the prepared NH 4 NO 3 salts (S1-S6), as analysed by the eight isotope laboratories and calibrated to Air-N 2 by analysis of IAEA and USGS standards, is indicated in   assumption is that the decrease in yield is at least partly caused by a "branching" side reaction, e.g. nitrogen gas (N 2 ) production, 53 which was observed to display higher δ 15 N(N 2 ) values. 1 We speculate that N 2 production has a minor effect on δ 15 N α , δ 15 N β and δ 15 N SP , but the effect is expected to depend on the timing of N 2 generation, which is not known.

| Consistency of isotopic composition of S1-N 2 O-S6-N 2 O
A general goal of the current project was to provide a link to the Air-      Tables S2 and S3 (supporting information). Table 7