Isotopic composition of atmospheric nitrate in a tropical marine boundary layer.

Long-term observations of the reactive chemical composition of the tropical marine boundary layer (MBL) are rare, despite its crucial role for the chemical stability of the atmosphere. Recent observations of reactive bromine species in the tropical MBL showed unexpectedly high levels that could potentially have an impact on the ozone budget. Uncertainties in the ozone budget are amplified by our poor understanding of the fate of NOx (= NO + NO2), particularly the importance of nighttime chemical NOx sinks. Here, we present year-round observations of the multiisotopic composition of atmospheric nitrate in the tropical MBL at the Cape Verde Atmospheric Observatory. We show that the observed oxygen isotope ratios of nitrate are compatible with nitrate formation chemistry, which includes the BrNO3 sink at a level of ca. 20 ± 10% of nitrate formation pathways. The results also suggest that the N2O5 pathway is a negligible NOx sink in this environment. Observations further indicate a possible link between the NO2/NOx ratio and the nitrogen isotopic content of nitrate in this low NOx environment, possibly reflecting the seasonal change in the photochemical equilibrium among NOx species. This study demonstrates the relevance of using the stable isotopes of oxygen and nitrogen of atmospheric nitrate in association with concentration measurements to identify and constrain chemical processes occurring in the MBL.

Laboratory study of nitrate photolysis in Antarctic snow. I. Observed quantum yield, domain of photolysis, and secondary chemistry.

Post-depositional processes alter nitrate concentration and nitrate isotopic composition in the top layers of snow at sites with low snow accumulation rates, such as Dome C, Antarctica. Available nitrate ice core records can provide input for studying past atmospheres and climate if such processes are understood. It has been shown that photolysis of nitrate in the snowpack plays a major role in nitrate loss and that the photolysis products have a significant influence on the local troposphere as well as on other species in the snow. Reported quantum yields for the main reaction spans orders of magnitude - apparently a result of whether nitrate is located at the air-ice interface or in the ice matrix - constituting the largest uncertainty in models of snowpack NOx emissions. Here, a laboratory study is presented that uses snow from Dome C and minimizes effects of desorption and recombination by flushing the snow during irradiation with UV light. A selection of UV filters allowed examination of the effects of the 200 and 305 nm absorption bands of nitrate. Nitrate concentration and photon flux were measured in the snow. The quantum yield for loss of nitrate was observed to decrease from 0.44 to 0.003 within what corresponds to days of UV exposure in Antarctica. The superposition of photolysis in two photochemical domains of nitrate in snow is proposed: one of photolabile nitrate, and one of buried nitrate. The difference lies in the ability of reaction products to escape the snow crystal, versus undergoing secondary (recombination) chemistry. Modeled NOx emissions may increase significantly above measured values due to the observed quantum yield in this study. The apparent quantum yield in the 200 nm band was found to be ∼1%, much lower than reported for aqueous chemistry. A companion paper presents an analysis of the change in isotopic composition of snowpack nitrate based on the same samples as in this study.

Laboratory study of nitrate photolysis in Antarctic snow. II. Isotopic effects and wavelength dependence.

Atmospheric nitrate is preserved in Antarctic snow firn and ice. However, at low snow accumulation sites, post-depositional processes induced by sunlight obscure its interpretation. The goal of these studies (see also Paper I by Meusinger et al. ["Laboratory study of nitrate photolysis in Antarctic snow. I. Observed quantum yield, domain of photolysis, and secondary chemistry," J. Chem. Phys. 140, 244305 (2014)]) is to characterize nitrate photochemistry and improve the interpretation of the nitrate ice core record. Naturally occurring stable isotopes in nitrate ((15)N, (17)O, and (18)O) provide additional information concerning post-depositional processes. Here, we present results from studies of the wavelength-dependent isotope effects from photolysis of nitrate in a matrix of natural snow. Snow from Dome C, Antarctica was irradiated in selected wavelength regions using a Xe UV lamp and filters. The irradiated snow was sampled and analyzed for nitrate concentration and isotopic composition (δ(15)N, δ(18)O, and Δ(17)O). From these measurements an average photolytic isotopic fractionation of (15)ɛ = (-15 ± 1.2)‰ was found for broadband Xe lamp photolysis. These results are due in part to excitation of the intense absorption band of nitrate around 200 nm in addition to the weaker band centered at 305 nm followed by photodissociation. An experiment with a filter blocking wavelengths shorter than 320 nm, approximating the actinic flux spectrum at Dome C, yielded a photolytic isotopic fractionation of (15)ɛ = (-47.9 ± 6.8)‰, in good agreement with fractionations determined by previous studies for the East Antarctic Plateau which range from -40 to -74.3‰. We describe a new semi-empirical zero point energy shift model used to derive the absorption cross sections of (14)NO3 (-) and (15)NO3 (-) in snow at a chosen temperature. The nitrogen isotopic fractionations obtained by applying this model under the experimental temperature as well as considering the shift in width and center well reproduced the values obtained in the laboratory study. These cross sections can be used in isotopic models to reproduce the stable isotopic composition of nitrate found in Antarctic snow profiles.

Analysis of oxygen-17 excess of nitrate and sulfate at sub-micromole levels using the pyrolysis method.

RATIONALE: The oxygen-17 excess (Δ(17)O) of nitrate and sulfate contains valuable information regarding their atmospheric formation pathways. However, the current pyrolysis method to measure Δ(17)O requires large sample amounts (>4 µmol for nitrate and >1 µmol for sulfate). We present a new approach employing a Gas Bench interface which cryofocuses O2 produced from sample pyrolysis, enabling the analysis of sub-micromole size samples. METHODS: Silver nitrate or sulfate at sub-micromole levels in a sample container was thermally decomposed to O2 and byproducts in a modified Temperature Conversion/Elemental Analyzer (TC/EA). Byproducts (mainly NO2 for silver nitrate and SO2 for silver sulfate) were removed in a liquid nitrogen trap and the sample O2 was carried by ultra-pure helium (He) gas to a Gas Bench II interface where it was cryofocused prior to entering an isotope ratio mass spectrometer. RESULTS: Analysis of the international nitrate reference material USGS35 (Δ(17)O = 21.6‰) within the size range of 300-1000 nmol O2 gave a mean Δ(17)O value of (21.6 ± 0.69) ‰ (mean ±1σ). Three inter-laboratory calibrated sulfate reference materials, Sulf-α, Sulf-ß and Sulf-ε, each within the size range of 180-1000 nmol O2, were analyzed and shown to possess mean Δ(17)O values of (0.9 ± 0.10)‰, (2.1 ± 0.25)‰ and (7.0 ± 0.63)‰, respectively. CONCLUSIONS: The analyses of nitrate and sulfate reference materials at sub-micromole levels gave Δ(17)O values consistent with their accepted values. This new approach of employing the Gas Bench to cryofocus O2 after the pyrolysis of AgNO3 and Ag2SO4 particularly benefits the effort of measuring Δ(17)O in sample types with a low abundance of nitrate and sulfate such as ice cores.

Measurement of the 17O-excess (Δ17O) of tropospheric ozone using a nitrite-coated filter.

RATIONALE: The (17)O-excess (Δ(17)O) of tropospheric ozone (O(3)) serves as a useful marker in studies of atmospheric oxidation pathways; however, due to the complexity and expense of currently available analytical techniques, no systematic sampling campaign has yet been undertaken and natural variations in Δ(17)O(O(3)) are therefore not well constrained. METHODS: The nitrite-coated filter method is a new technique for O(3) isotope analysis that employs the aqueous phase NO(2)(-) + O(3) → NO(3)(-) + O(2) reaction to obtain quantitative information on O(3) via the oxygen atom transfer to nitrate (NO(3)(-)). The triple-oxygen isotope analysis of the NO(3)(-) produced during this reaction, achieved in this study using the bacterial denitrifier method followed by isotope-ratio mass spectrometry (IRMS), directly yields the Δ(17)O value transferred from O(3). This isotope transfer process was investigated in a series of vacuum-line experiments, which were conducted by exposing coated filters to O(3) of various known Δ(17)O values and then determining the isotopic composition of the NO(3)(-) produced on the filter. RESULTS: The isotope transfer experiments revealed a strong linear correlation between the Δ(17)O of the O(3) produced and that of the oxygen atom transferred to NO(3)(-), with a slope of 1.55 for samples with bulk Δ(17)O(O(3)) values in the atmospheric range (20-40‰). This finding is in agreement with theoretical postulates that place the (17) O-excess on only the terminal oxygen atoms of ozone. Ambient measurements yield average Δ(17)O(O(3))(bulk) values in agreement with previous studies (22.9 ± 1.9‰). CONCLUSIONS: The nitrite-coated filter technique is a sufficiently robust, field-deployable method for the determination of the triple-oxygen isotopic composition of tropospheric O(3). Further ambient measurements will undoubtedly lead to an improved quantitative view of natural Δ(17)O(O(3)) variation and transfer in the atmosphere.

Oxygen isotope exchange with quartz during pyrolysis of silver sulfate and silver nitrate.

RATIONALE: Triple oxygen isotopes of sulfate and nitrate are useful metrics for the chemistry of their formation. Existing measurement methods, however, do not account for oxygen atom exchange with quartz during the thermal decomposition of sulfate. We present evidence for oxygen atom exchange, a simple modification to prevent exchange, and a correction for previous measurements. METHODS: Silver sulfates and silver nitrates with excess (17)O were thermally decomposed in quartz and gold (for sulfate) and quartz and silver (for nitrate) sample containers to O(2) and byproducts in a modified Temperature Conversion/Elemental Analyzer (TC/EA). Helium carries O(2) through purification for isotope-ratio analysis of the three isotopes of oxygen in a Finnigan MAT253 isotope ratio mass spectrometer. RESULTS: The Δ(17)O results show clear oxygen atom exchange from non-zero (17)O-excess reference materials to zero (17)O-excess quartz cup sample containers. Quartz sample containers lower the Δ(17)O values of designer sulfate reference materials and USGS35 nitrate by 15% relative to gold or silver sample containers for quantities of 2-10 µmol O(2). CONCLUSIONS: Previous Δ(17)O measurements of sulfate that rely on pyrolysis in a quartz cup have been affected by oxygen exchange. These previous results can be corrected using a simple linear equation (Δ(17)O(gold) = Δ(17)O(quartz) * 1.14 + 0.06). Future pyrolysis of silver sulfate should be conducted in gold capsules or corrected to data obtained from gold capsules to avoid obtaining oxygen isotope exchange-affected data.