Introduction

There are five species of interest when analyzing N₂O isotopes: the abundant isotopomer ¹⁴N¹⁴N¹⁶O and the rare isotopomers ¹⁴N¹⁵N¹⁶O, ¹⁵N¹⁴N¹⁶O, ¹⁴N¹⁴N¹⁷O and ¹⁴N¹⁴N¹⁸O. The two ¹⁵N species are indistinguishable with current mass spectrometric techniques. The two rare oxygen species have in the past been assumed to be mass-dependently related although it has recently been shown that there is a slight mass-independent enrichment of ¹⁷O in tropospheric samples [Cliff and Thiemens, 1997] which is as yet unaccounted for. Isotopic values are typically reported as ratios of the heavy-to-light species relative to those in a standard such that d = [(R_samp/R_std) ­1] x 1000 (expressed in units per mil).

Because the isotopomers of interest are isobaric with those of CO₂ (masses 44, 45, and 46) and since CO₂ in natural environments is generally ~1000 times more abundant, traditional methods of isotopic analysis employed techniques involving selective decomposition of N₂O and subsequent analysis of products. For this reason, the isotopic standards for N₂O have been atmospheric N₂ for ¹⁵N and either atmospheric O₂ or Standard Mean Ocean Water for the oxygen isotopes. (Atmospheric O₂ has become the de facto standard for the oxygen isotopes.) In the case of current analytical techniques (with the exception of ¹⁷O analyses), N₂O is now routinely separated from CO₂ and fed directly into a mass spectrometer. It is then referenced to a nitrous oxide gas standard with known isotopic values relative to N₂ and O₂.

Nitrous oxide is a trace gas that is produced during microbial energy exchange reactions involving both reduced (NH₃) and oxidized (NO₃^­) forms of nitrogen. A portion of the N₂O that is produced escapes to the troposphere where it is chemically inert. The commonly accepted model holds that nitrous oxide ascends to the stratosphere where it is photolyzed by ultraviolet radiation (90% of the total loss); oxidized by excited atomic oxygen (10% of loss); or returned to the troposphere during stratosphere/troposphere exchange processes. The bulk of the N₂O photolysis, as demonstrated in Figure 1, occurs on the shoulder of the cross section spectrum, between the Schumann-Runge bands and Herzberg continuum of O₂ absorption, rather than at the peak. While in the atmosphere, N₂O actively absorbs infrared radiation and thereby contributes to greenhouse warming. A portion of the N₂O destroyed by the reac-reactants under equilibrium conditions. Of the three important biologically mediated greenhouse gases, our understanding of the isotopic budget of nitrous oxide lags far behind that of carbon dioxide and methane. This is due in part to problems inherent in collection and analytical techniques which hamper our ability to make measurements of very high precision. It is also due to the fact that a limited data base and a wide range of observed isotopic values for each of the major natural sources has made it difficult to assign unique values to each of the source terms.

Figure 1: High resolution N2O cross section from Yoshino et al. [1984] and the actinic flux as measured at 40 km by Herman and Mentall [1982] (0.21 nm resolution).

The isotopic signature of tropospheric N₂O is commonly reported as d¹⁵N = 7.0 and d¹⁸O = 20.7 [Kim and Craig, 1990] although inter-laboratory averages vary by 3 and 4 for d¹⁵N and d¹⁸O, respectively [see Moore, 1974; Yoshida and Matsuo, 1983; Yoshida et al., 1984; Wahlen and Yoshinari, 1985; Kim and Craig, 1990; Cliff and Thiemens, 1997; Rahn and Wahlen, 1997]. The measured isotopic signature of N₂O emitted from terrestrial environments is nearly everywhere depleted relative to the tropospheric average. The averages of known measurements from terrestrial sources, including fertilized lands, are ­14.6 ± 11.9 for d¹⁵N; and 8.2 ± 7.1 for d¹⁸O [Kim and Craig, 1993; Casciotti, 1997; T. Pérez, pers. comm]. Oceanic surface waters are observed to be typically depleted in both ¹⁵N and ¹⁸O [Yoshida et al., 1984; Kim and Craig, 1990, Dore et al., 1998] although it is proposed that in certain upwelling areas there may be isotopically enriched deep waters introduced to the surface [Yoshinari et al., 1997]. The observation that the source terms are depleted relative to the tropospheric reservoir led Kim and Craig [1993] to propose that the enrichments which they observed in the stratosphere might provide the necessary balance. A conundrum resulted when it was shown by Johnston et al. [1995] that the fractionations associated with the major N₂O loss terms, photolysis and photooxidation, were not great enough to account for the observed stratospheric values. Rahn and Wahlen [1997] subsequently verified that stratospheric N₂O is indeed enriched in the heavy stable isotopes of both N and O (see Figure 2). These stratospheric results have led to speculation that the anomalous enrichment might provide evidence of novel excited state photochemical sources of nitrous oxide [McElroy and Jones, 1996; Prasad, 1997; Prasad et al., 1997; Zipf and Prasad, 1998]. In particular, Zipf and Prasad [1998] suggest that highly vibrationally excited O₃ could react with N₂ to form N₂O and O₂ and that this reaction could account for as much as 8% of the global source of nitrous oxide. Such photochemical sources would significantly alter our understanding of the geochemical cycle of N₂O. The results of Rahn and Wahlen [1997], however, are in good agreement with a Rayleigh distillation loss model indicating that no significant source products are interfering with the isotopic signature of the destruction mechanism. Source processes, should they tion with O(¹D) provides the principle natural source of NO, which initiates the catalytic NO_x cycling of stratospheric ozone. The current tropospheric concentration, which is increasing at a rate of ~0.25% per year, is about 313 ppbv and the estimated atmospheric lifetime is approximately 120 years. Because of its influence on the Earth's radiative budget and its increasing concentration, nitrous oxide has been selected as one of the six gases slated for regulation by the Kyoto Protocol of 1997.

Figure 2: Lower stratosphere (between 14 and 19 km) isotopic enrichments of 15N and 18O. The results compare favorably with a Rayleigh distillation model where R = Rof(a-1) : with R and Ro equal to the residual and initial heavy-to-light isotope ratios; f, the fraction of tropospheric N2O remaining; and a, the ratio of heavy-to-light destruction rates. Note that the regressions pass through tropospheric values at ln(f) = 0. A single sample from 22.4 km (not shown) shows reasonable agreement for 18O and an increased enrichment for 15N indicating the possibility of competing processes at higher altitudes. The figure is based on Figure 1 of Rahn and Wahlen [1999].

Despite the importance of nitrous oxide and its incorporation in the Kyoto Protocol, its global budget is poorly characterized. Estimates of the natural sources of N₂O range from 1 to 5 Tg N/yr for oceanic sources and 3.3 to 9.7 Tg N/yr from tropical and temperate soils [IPCC, 1995]. The atmospheric increase is considered to arise primarily from application of fertilizers to cultivated soils but animal waste, biomass burning, fuel combustion, and industrial processes also contribute. The estimated range of the sum of the anthropogenic sources is 3.7 to 7.7 Tg N/yr [IPCC, 1995]. In an effort to reduce the error in these estimates, investigations of the stable isotopic signatures of the various sources and sinks have been carried out by several research groups [see Moore, 1974; Yoshida and Matsuo, 1983; Wahlen and Yoshinari, 1985; Yoshinari and Wahlen, 1985; Kim and Craig, 1990; Kim and Craig, 1993; Yoshinari et al., 1997; Naqvi et al., 1998; Dore et al., 1998].

Isotopic considerations

The stable isotopic composition of atmospheric trace gases provides information about their origin and fate that cannot be determined from concentration measurements alone. Biological source and loss processes, such as bacterial production of CH₄ or photosynthetic consumption of CO₂, are typically accompanied by isotopic selectivity associated with the kinetics of bond formation and destruction. Thermodynamic considerations also predict isotopic differentiation between phases and/or exist, would therefore have to be able to mimic the Rayleigh distillation model, or be associated with negligible fractionation, or represent a very minor portion of the global budget.

Photolytic fractionation

An alternative explanation for the isotopic enrichment of stratospheric N₂O has been provided by Yung and Miller [1997]. They have proposed a wavelength-dependent mechanism for the photolytic fractionation of N₂O based on subtle shifts in the zero point energy with isotopic substitution. The wavelength dependence allows for the minimal fractionation observed by Johnston et al. [1995] near the peak of the absorption cross section at 184.9 nm yet predicts increasing fractionation at longer wavelengths where the bulk of the stratospheric N₂O photolysis takes place. This principle is demonstrated graphically in Figure 3 using the absorption cross section spectral function recommended by Selwyn et al. [1977]. The curve representing the ¹⁸O substituted species (dashed curve) is slightly blue shifted, by ­27.5 cm^­1 [Jung and Miller, 1997], relative to the normal curve (both calculated at 300 K). Cross sections are nearly equal in the region of the curve crossing at the absorption peak but a clear separation is observed on both shoulders. The inset plot of Figure 3 details the highlighted section on the higher wavelength shoulder. Pointed out are the cross sections at 185 nm for the normal curve (14.00 x 10^-20 cm²) and the blue shifted N₂¹⁸O curve (13.96 x 10^-20 cm²). Analogous to determining the kinetic fractionation for a chemical reaction, the photolytic fractionation factor will be equal to the ratio of the heavy to light cross sections or a=0.9971. When expressed as an enrichment factor, where e = 1000(a-1), e₁₈₅= -2.9. The theoretical enrichment factors can be calculated and plotted as a function of wavelength. As can be seen in Figure 4, the wavelength dependent enrichment factor, e_l, is initially positive at shorter wavelengths, passes through zero at the cross section maximum, and gets progressively more negative with increasing wavelength.

Similar constructs can be developed for each of the remaining isotopic species, all of which exhibit similar behavior with varying magnitude. However, the asymmetry of nitrous oxide leads to complications in the ¹⁵N analyses because the two isotopomers, ¹⁴N¹⁵NO and ¹⁵N¹⁴NO, have different predicted blue shifts. Yung and Miller [1997] account for this complexity by averaging the predicted enrichment of the two ¹⁵N species. If the averaged ¹⁵N value and the ¹⁸O value are taken as a ratio , a fairly uniform value between 1.1 and 1.2 is calculated over all wavelengths. This is in good agreement with the ratio observed in the stratosphere by Rahn and Wahlen [1997] where

In reality, the situation is complicated even further by vibrational structure in the absorption continuum. If we apply the same treatment as described above to the cross section curves of Selwyn and Johnston [1981], we see that there are multiple curve crossings causing e_l to change sign several times over the cross section spectrum (Figure 5). Included in Figure 5 are the results of Rahn et al. [1998] for laser induced photolysis at two discrete wavelengths. The results compare favorably with the predicted e_l value at 193 nm but the data of Selwyn and Johnston [1981] do not extend to wavelengths greater than 197 nm making a comparison of the 207 nm data impossible.

The results shown in Figure 5 are approximately double that predicted by the theory of Yung and Miller [1997]. While the absolute magnitudes of the observed and predicted fractionation are significantly different, the general concept of enrichment being caused by spectral shifts induced by isotopic substitution is still a valid and likely mechanism. A more rigorous treatment of the model, including non-Born-Op-penheimer effects and dipole moment surface variations, might yield better quantitative agreement.

The laboratory results are also significantly greater, on average, than the values observed in the stratosphere by Rahn and Wahlen [1997]. It has been shown, however, that the standard Rayleigh fractionation model scales with (a^1/2 ­ 1) rather than (a - 1) in a purely diffusive regime [Eriksson, 1965]. If the integrated fractionation factors under stratospheric conditions can be determined, this result may prove useful in models dealing with stratospheric transport processes and troposphere/stratosphere exchange.

Conclusions

Since photolysis is the most important sink of atmospheric nitrous oxide, any attempts at constructing a global N₂O budget incorporating isotopic results must take into account this photolytic fractionation. In order to determine the integrated stratospheric fractionation, enrichment factors at high spectral resolution will have to be determined and convolved with models of stratospheric actinic flux. This will be an essential parameter if we are to advance our understanding of the global nitrous oxide isotopic budget and could also provide a new tool for examining issues of troposphere/stratosphere exchange.

Fractionation associated with the minor sink, reaction with excited atomic oxygen, must also be taken into account. The fractionation due to this reaction has been determined for the oxygen species, ¹⁷O and ¹⁸O, and has been shown to be mass dependent with = -6 ± 1 [Johnston et al., 1995]. The fractionation due to photooxidation for the N species has yet to be determined. Finally, we note that the asymmetry of N₂O presents a unique opportunity to investigate an additional set of parameters with which we can constrain the N₂O budget. The fractionation for the two ¹⁵N isotopomers is predicted to be significantly different according to the theory of Yung and Miller [1997], and should be detectable with advanced optical techniques. It will be interesting to determine someday if the biologically mediated source terms have unique ¹⁵N position signatures as well.

References

1. Casciotti, K., Rahn, T., Wahlen, M., Stable isotopes of N and O in nitrous oxide emissions from fertilized soil (abstract), Fall Meeting of Am. Geophys. U., San Francisco, EOS Supplement, F, 1997.
2. Cliff, S. S., and M. H. Thiemens, The ¹⁸O/¹⁶O and ¹⁷O/¹⁶O ratios in atmospheric nitrous oxide: A mass-independent anomaly. Science, 278, 1774-1776, 1997.
3. Dore, J. E., B. Popp, D. Karl, F. Sansone, A large source of atmospheric nitrous oxide from subtropical North Pacific surface waters, Nature, 396, 63-66, 1998.
4. Eriksson, E., Deuterium and oxygen-18 in precipitation and other natural waters: Some theoretical considerations, Tellus, XXVI, 498-512, 1965.
5. Herman, J., and J. Mentall, The direct and scattered solar flux within the stratosphere, J. Geophys. Res., 87, 1319-1330, 1982.
6. IPCC, (Intergovernmental Panel on Climate Change), Climate change, 1994 : radiative forcing of climate change and an evaluation of the IPCC IS92 emission scenarios, J.T. Houghton, [et al.], Ed., Cambridge [England]; New York: Cambridge University Press, 1985.
7. Johnston, J. C., S. Cliff, and M. Thiemens, Measurement of multioxygen isotopic (d¹⁸O and d¹⁷O) fractionation factors in the stratospheric sink reactions of nitrous oxide, J. Geophys. Res., 100, 16801-16804, 1995.
8. Kim, K.-R., and H. Craig, Two isotope characterization of N₂O in the Pacific Ocean and constraints on its origin in deep water, Nature, 347, 58-61, 1990.
9. Kim, K.-R., and H. Craig, Nitrogen-15 and oxygen-18 characteristics of nitrous oxide: A global perspective, Science, 262, 1855-1857, 1993.
10. McElroy, M. B., and D. B. A. Jones, Evidence for an additional source of atmospheric N₂O, Glob. Biogeochem. Cycles, 10, 651-659, 1996.
11. Moore, H., Isotopic measurement of atmospheric nitrogen compounds, Tellus, XXVI, 169-174, 1974.
12. Prasad, S. S., Potential atmospheric sources and sinks of nitrous oxide 2: Possibilities from excited O₂, ''embryonic'' O₃, and optically pumped excited O₃, J. Geophys. Res., 102, 21527-21536, 1997.
13. Prasad, S. S., E. C. Zipf, and X. P. Zhao, Potential atmospheric sources and sinks of nitrous oxide 3: Consistency with the observed distributions of the mixing ratios, J. Geophys. Res., 102, 21537-21541, 1997.
14. Rahn, T., and M. Wahlen, Stable isotope enrichment in stratospheric nitrous oxide, Science, 278, 1776-1778, 1997.
15. Rahn, T, H. Zhang, M. Wahlen, G. Blake, Stable isotope fractionation during ultraviolet photolysis of N₂O, Geophys. Res. Lett., 25 , 4489-4492, 1998.
16. Selwyn, G., J. Podolske, H.S. Johnston, Nitrous oxide ultraviolet absorption spectrum at stratospheric tempertures, Geophys. Res. Lett., 4, 427-430, 1977.
17. Selwyn, G., and H. S. Johnston, Ultraviolet absorption spectrum of nitrous oxide as function of temperature and isotopic substitution, J. Chem. Phys., 74, 3791-3803, 1981.
18. Wahlen, M. and T. Yoshinari, Oxygen isotope ratios in N₂O from different environments, Nature, 313, 780-782, 1985.
19. Yoshida, N., and S. Matsuo, Nitrogen isotope ratio of atmospheric N₂O as a key to the global cycle of N₂O, Geochemical Journal, 17, 231-239, 1983.
20. Yoshida, N., et al., ¹⁵N/¹⁴N ratio of dissolved N₂O in the eastern tropical Pacific Ocean, Nature, 307, 442-444, 1984.
21. Yoshinari, T., and M. Wahlen, Oxygen isotope ratios in N₂O from nitrification at a wastewater treatment facility, Nature, 317, 349-350, 1985.
22. Yoshinari, T., et al., Nitrogen and oxygen isotopic composition of N₂O from suboxic waters of the eastern tropical North Pacific and the Arabian Sea — measurement by continuous-flow isotope-ratio monitoring, Marine Chem., 56, 253-264 1997.
23. Yung, Y. L., and C. E. Miller, Isotopic fractionation of stratospheric nitrous oxide via photolysis, Science, 278, 1778-80, 1997.
24. Zipf, E., and S. Prasad, Experimental evidence that excited ozone is a source of nitrous oxide, Geophys. Res. Lett., 25, 4333-4336, 1998.