Introduction

There are four naturally occurring stable isotopes of sulfur ³²S, ³³S, ³⁴S and ³⁶S with approximate abundances of 95.02, 0.75, 4.21 and 0.02%, respectively (MacNamara and Thode, 1950). Isotopic variations are almost always considered in terms of the ratio of the two most abundant isotopes ³⁴S/³²S. Isotope effects are usually small and fractional differences in isotope ratios (dvalues) are normally expressed in parts per thousand () relative to a standard:

The generally accepted standard is a meteoritic iron sulfide, the Canyon Diablo Troilite. There is evidence that the ratio of ³⁴S/³²S in this type of meteorite is the primordial value for sulfur on earth, and is therefore a natural baseline to which terrestrial variations can be related.

The ³⁴S /³²S ratio is measured by stable isotope mass spectrometry most commonly using sulfur dioxide, although sulfur hexafluoride is sometimes used. Traditional off-line preparation methods for sulfur dioxide are time consuming and require rather large amounts of sulfur (~300µg S), which have placed limitations on their usefulness for atmospheric samples. However, recent use of an elemental analyser coupled to a mass spectrometer (e.g., Giesemann et al., 1994, Patris et al., 1998) is reducing both sample size (to ~20µg S) and preparation times, although some loss of precision is a potential problem.

Sources of sulfur to the atmosphere have a wide range in d³⁴S values and this is reflected in the d³⁴S of sulfur sampled in the atmosphere (Figure 1). Although the range of d³⁴S values found for samples of fossil fuels is large, the range of d³⁴S observed in the flue gases of fuels actually combusted can be very much smaller. For example Newman and Forrest (1991) found a range in d³⁴S values of only ­4 to +8 for flue gases from coal and oil fired power plants in the northeastern USA.

Figure 1. Variations of d34S values for different sources of sulfur to the atmosphere (after Thode, 1991), and some measurements from atmospheric aerosol samples. [Long Island: Newman and Forrest, 1991; Mace Head: McArdle and Liss, 1995; Pacific Ocean: Calhoun et al., 1995].

In addition, the ratio of the two isotopes can be altered (fractionated) during biological, physical and chemical transformations due to the difference in mass between the isotopes. In a unidirectional chemical reaction a kinetic isotope effect occurs when the lighter isotope reacts faster, so at any instant in time the products will be lighter than the reactants. However, if all the reactants are transformed there will be no overall fractionation between reactants and products. Fractionation during equilibrium reactions may lead to an enrichment in the products of either the heavier or lighter isotope.

A process by which sulfur isotopes are commonly altered in nature is by biological transformation of the sulfur, either by oxidation or reduction. There are two biochemical pathways to sulfate reduction: assimilatory sulfate reduction, which is thought to result in little or no isotopic fractionation, and dissimilatory sulfate reduction usually associated with large fractionations. Bacterial oxidation of reduced inorganic sulfur compounds can also result in significant fractionations.

Applications

Sulfur isotopes can be used to trace the emissions of point sources if they have sufficiently different d³⁴S values compared to environmental receptors. For example, the emissions from sour gas processing plants in Canada are significantly enriched in ³⁴S and have provided an effective tool for tracing the fate of these emissions in surrounding ecosystems [Krouse, 1991].

On a larger scale sulfur isotopes can be used to look at historical changes in the amount of sulfur from human sources that has been taken up into the terrestrial biosphere and soil. Figure 2 shows data from a study by Zhao et al. [1998] of soil and herbage samples collected in the UK over the past 140 years. The variations in sulfur content of the plants reflect changing anthropogenic emissions of SO₂ in the UK. The d³⁴S of the plants has an opposite trend indicating that the additional sulfur has a low d³⁴S, consistent with that expected from the combustion of coal. The authors estimate that anthropogenic S contributed up to 50% of the herbage uptake at the peak of SO₂ emissions (in the early 1970s).

Figure 2 (a). Sulfur content in grassland plants and UK sulfur dioxide emissions from 1860 to 1996. (b) Herbage d34S variations from 1860 to 1996 [from Zhao et al., 1998].

Isotopic signatures can also be used for quantitative source apportionment if the isotopic compositions of atmospheric sulfur from different sources are known and are sufficiently different from each other. Perhaps one of the most interesting applications of sulfur isotopes is for assessing the magnitude of the marine biogenic component (i.e., from the oxidation of dimethylsulfide (DMS) produced by phytoplankton) in remote and relatively remote locations, where anthropogenic sources have been well mixed resulting in a reasonably homogeneous anthropogenic d³⁴S value. In order to do this one needs a d³⁴S value for DMS derived sulfur. However, as yet, there are no direct measurements of the isotopic signature of DMS in the atmosphere or of sulfate that is unequivocally derived from DMS. A theoretical consideration of the production pathway for atmospheric sulfate from seawater via oxidation of DMS suggests fractionations of +1 to ­7 relative to seawater, i.e., d³⁴S of +13 to +21. [Calhoun et al., 1991]. This is consistent with d³⁴S values observed in atmospheric samples from remote locations.

A study of aerosol samples collected at Mace Head on the west coast of Eire and Ny Ålesund, Spitsbergen suggested that the signature of sulfate derived from DMS is probably at the higher end of the predicted range [McArdle and Liss, 1995]. At the latitudes of the sampling sites in that study DMS production is strongly seasonal with little being produced during the winter, confirmed by the very low observed MSA (methanesulfonic acid, another oxidation product of DMS) concentrations compared with the spring and summer months. It was assumed that winter samples contained no DMS derived sulfur but were a mixture of sea-salt sulfate and terrestrial (mostly anthropogenic) sulfate. The contribution of sea-salt sulfate can be calculated assuming that sodium is a conservative tracer for oceanic sulfate. Figures 3 (a) and (b) show d³⁴S values plotted against % sea-salt sulfate for winter and summer. In winter the data are distributed along a two-source mixing line, with the spread giving an indication of the variability in the terrestrial signature at these two sites.

Figure 3. Plot of d34S against percentage sea-salt sulfate for (a) winter and (b) summer aerosol samples collected at Mace Head (solid dots) and Ny Ålesund (open circles). The linear regression line is for the Mace Head winter data and has intercepts of +4.4 and +20.1 at 0 and 100 % sea-salt sulfate, respectively (from McArdle and Liss, 1995).

In summer the distribution of data is very different with points lying well above the winter mixing line. This indicates the presence of an isotopically heavy source of sulfur during the summer which was assumed to be derived from the oxidation of DMS with an estimated d³⁴S of +22 . In this study the biogenic sulfur was found to make up ~30% of the sampled sulfur for the spring and summer months. An isotopic study of aerosol sulfate at Alert, Canadian Arctic also found biogenic contributions of 25-30% during the summer [Li and Barrie, 1993].

In studies such as this the terrestrial/anthropogenic signature needs to be constrained as tightly as possible for each location as biogenic contributions are often relatively small, particularly in the Northern Hemisphere. However, there seems to be rather little variation in d³⁴S values across Europe and N. America with remarkably consistent terrestrial d³⁴S values found in other studies around the N. Atlantic [e.g., McArdle et al., 1998, Wadleigh et al., 1996, Newman and Forrest, 1991, for sites in Wales, Nova Scotia and Long Island, respectively].

A number of studies have utilized isotopic fractionations to examine atmospheric processes, for example determining the relative importance of homogeneous (gas phase) versus heterogeneous (aqueous phase, in cloud droplets or on the surface of wet aerosols) oxidation of sulfur dioxide. The fractionation associated with a homogeneous oxidation pathway is a kinetic effect whereby the lighter isotope reacts faster, resulting in sulfate isotopically lighter than the initial SO₂. Estimates of this fractionation range from 4 to 9 lighter [Saltzman et al., 1983; Tanaka et al., 1994]. The heterogeneous pathway, on the other hand, is expected to result in sulfate isotopically heavier than the SO₂. However, the exact value of the total fractionation is unknown as several reactions are involved, with sulfate perhaps up to 16.5 heavier [Tanaka et al., 1994]. But, field studies have only found small fractionations between gas and aerosol or rain in the range 0 to +4, leading authors to postulate that both processes are important, with one fractionation tending to cancel out the other [Saltzman et al., 1983].

The development of new methods that reduce sample size expands the potential applications of this powerful technique, allowing for example better time resolution of samples, especially for remote southern hemisphere locations and ice cores. Also, there are good possibilities of combining stable isotope measurements of sulfur with those of other elements, such as oxygen or nitrogen, to provide more information about sources and environmental processes.

References

1. Calhoun, J.A., T.S. Bates, and R.J. Charlson, Sulfur isotope measurements of submicrometer sulfate aerosol particles over the Pacific Ocean, Geophys. Res. Lett. 18, 1877-1880, 1991.
2. Giesemann, A., H.-J. Jäger, A.L. Norman, H.R. Krouse and W.A. Brand, On-line sulfur-isotope determination using an elemental analyzer coupled to a mass spectrometer, Anal. Chem., 66, 2816-2819, 1994.
3. Krouse, H.R., Sulphur isotope tracing of the fate of emissions from sour gas processing in Alberta, Canada, In Stable Isotopes: Natural and Anthropogenic Sulphur in the Environment. SCOPE 43. edited by H.R. Krouse, and V.A. Grinenko, pp 309-326, John Wiley, New York, 1991.
4. Li, S.-M. and L.A. Barrie, Biogenic sulfur aerosols in the Arctic troposphere: 1. Contributions to total sulfate. J. Geophys. Res., 98, 20,613-20,622, 1993.
5. MacNamara, J. and H.G. Thode, Comparison of the isotopic composition of terrestrial and meteoritic sulfur. The Physical Rev., 78, 307-308, 1950.
6. McArdle, N.C. and P.S. Liss, Isotopes and atmospheric sulphur. Atmos. Environ., 29, 2553-2556, 1995.
7. McArdle, N., P. Liss, and P. Dennis, An isotopic study of atmospheric sulphur at three sites in Wales and at Mace Head, Eire, J. Geophys. Res., 103, 31,079-31,094, 1998.
8. Newman, L. and J. Forrest, Sulphur isotope measurements relevant to power plant emissions in the northeastern United States. In Stable Isotopes: Natural and Anthropogenic Sulphur in the Environment. SCOPE 43. edited by H.R. Krouse, and V.A. Grinenko, pp 331-343, John Wiley, New York, 1991.
9. Patris, N., J. Jouzel, and R. Delmas, Isotopic composition of sulfur traces in Polar and Alpine ice cores. Abstract, IGAC Science Conference, Seattle, August 1998.
10. Saltzman, E.S., G. Brass, and D. Price, The mechanism of sulfate aerosol formation: Chemical and sulfur isotopic evidence. Geophys. Res. Lett., 10, 513-516, 1983.
11. Tanaka, N., D.M. Rye, Y. Xiao, and A.C. Lasaga, Use of stable sulfur isotope systematics for evaluating oxidation reaction pathways and in-cloud-scavenging of sulfur dioxide in the atmosphere. Geophys. Res. Lett., 21, 1519-1522, 1994.
12. Thode, H.G., Sulphur isotopes in nature and the environment: An overview. In Stable Isotopes: Natural and Anthropogenic Sulphur in the Environment. SCOPE 43. edited by H.R. Krouse, and V.A. Grinenko, pp 1-26, John Wiley, New York, 1991.
13. Wadleigh, M.A., H.P. Schwarcz, and J.R. Kramer, Isotopic evidence for the origin of sulphate in coastal rain. Tellus Ser.B, 48, 44-59, 1996.
14. Zhao, F.J., B.Spiro, P.R. Poulton, and S.P. McGrath, Use of sulfur isotope ratios to determine anthropogenic sulfur signals in a grassland ecosystem, Environ. Sci. Tech., 32, 2288-2291, 1998.