A method for nitrite removal in nitrate N and O isotope analyses

We describe a method to remove nitrite from seawater and freshwater samples to determine the nitrogen (N) and oxygen (O) isotopic composition of nitrate in aqueous samples (15N/14N and 18O/16O, respectively). This method is simple, inexpensive, and effective in the removal of nitrite without compromising the natural N and O isotopic composition of nitrate. Moreover, the method is nontoxic and compatible with the “denitrifier method” for nitrate N and O isotope analysis. Nitrite is removed from solution by reduction to nitric oxide (NO) gas using ascorbic acid at a pH of ~3.5. The NO produced is continually degassed with an inert gas during the reaction, preventing O2 from reacting with NO to form new nitrate. Nitrate N and O isotope ratios were measured with the denitrifier method. The precision of isotope measurements of nitrate by the denitrifier method after nitrite removal averaged ±0.34‰ for 15N/14N and ±0.39‰ for 18O/16O. Acknowledgments We thank Greg Cane for technical assistance with nitrate isotope measurements. This work was funded by post-graduate NSERC award (National Science and Engineering Research Council of Canada, to J.G.), the US NSF (through grant OCE-0447570 to D.M.S.), and a CEBIC seed grant (Center for Environmental Bioinorganic Chemistry). We thank Dr. Karen Casciotti for important insights during development of the method, as well as two anonymous reviewers who helped improve the manuscript significantly. Limnol. Oceanogr.: Methods 4, 2006, 205–212 © 2006, by the American Society of Limnology and Oceanography, Inc. LIMNOLOGY and OCEANOGRAPHY: METHODS topic fractionation imparted on nitrate by denitrification therefore require that interference from nitrite be eliminated. A number of published methods allow for the removal of nitrite without affecting nitrate concentrations. Binding of nitrite to sulfamic acid has been used previously to measure N isotope ratios of nitrate (Wu et al. 1997). However, sulfamic acid is a potent antibiotic and thus incompatible with the denitrifier method. Binding of nitrite to iodide (Garside 1982), a method commonly used to detect low concentrations of nitrite, was also deemed problematic to measure nitrate isotopes, in part because high concentrations of iodide are likely to be toxic to bacteria, and also because the method requires very low pH, which can lead to oxygen atom exchange between nitrate and water (Bunton et al. 1952). We initially used hydroxylamine to remove nitrite from our samples, the product being N2O (BothnerBy and Friedman 1952). Though this reagent is nontoxic, any hydroxylamine remaining in solution competed with denitrifiers for the nitrite being produced from the sample nitrate, disturbing the isotopic relationships between a nitrate sample and its product N2O. We have developed an ascorbate-based method to remove nitrite from both freshwater and seawater samples that imparts no change in the concentration or the N and O isotopic composition of coincident nitrate. The procedure relies on the capacity of ascorbate to reduce nitrite to nitric oxide (NO) gas in mildly acidic solution at room temperature. The method is nontoxic to the bacteria used in the denitrifier method of Sigman et al. (2001) and Casciotti et al. (2002) and is cost-effective. The method presented here allows for precise and accurate quantification of nitrate N and O isotopic ratios in aqueous samples where nitrite concentrations are significant. Materials and procedures Reduction of nitrous acid by ascorbate—The pKa1 of ascorbic acid (AscH2) is 4.2, around which pH the concentration of nitrous acid (HNO2) becomes sufficiently high to engender its spontaneous reduction by ascorbate (AscH–), generating NO (nitric oxide) gas (Kanda and Taira 2003): AscH2 ↔ AscH – + H+ (pKa1 4.2) (1) HNO2 ↔ NO2 – + H+ (pKa 3.3) (2) AscH– + 2HNO2 → DHA (dehydroxyascorbic acid) + 2NO· + 2H2O (3) Whereas ascorbate readily reduces nitrous acid, it does not reduce nitrate. Thus, nitrite can be selectively removed from aqueous samples with an ascorbic acid addition sufficient to bring the solution down to a pH of around 3.5. However, diatomic oxygen can readily react with the NO free radical and oxidize it to nitrogen dioxide (NO2) (Equation 4). The latter can then react with the ascorbate anion to form nitrite (Equation 5) or with water to form both nitrite and nitrate (Equation 6): 2 NO· + O2 → 2 NO2· (4) NO2· + AscH – → NO2 – + Asc·– + H+ (5) 2 NO2· + H2O → HNO3 + HNO2 (6) To avoid these reactions, nitric oxide gas is removed continually throughout the reaction by bubbling the solution with an inert gas, maintaining a low oxygen concentration in the sample and sweeping away product NO before it reacts with any oxygen that is present. Methodology—Nitrate and nitrite concentrations were measured by conversion to NO followed by chemiluminescence detection (Braman and Hendrix 1989) on an Antek 1750 nitrate/nitrite analyzer. Our limit of detection was ≤ 20 nM nitrate or nitrite. Nitrite was also quantified colorimetrically by reaction with Greiss reagents (sulfanilamide and N-1naphthyleneethylenediamine [NNED]) and measurement of absorbance at 543 nm. Before use, glassware and plasticware were acid washed (10% HCl) and rinsed with milli-Q water (Millipore). The culture medium samples analyzed in this study consisted of synthetic ocean water (Price et al. 1988/89) made from milli-Q water. This mixture did not contain any detectable nitrate or nitrite. Freshwater samples refer to milli-Q water that was not amended with salts. A nitrate consumption experiment with the denitrifying bacterium Pseudomonas aureofaciens (ATCC 13985, recently reclassified as a strain of P. chlororaphis) was used to test nitrite removal by the present method. The strain was grown in artificial seawater medium with ~260 μM initial nitrate according to a procedure outlined previously (Granger and Ward 2003). The 15N/14N and 18O/16O of nitrate and/or nitrite were determined following the denitrifier method (Casciotti et al. 2002; Sigman et al. 2001). Isotope ratios are reported using delta (δ) notation in units of per mil (‰): δNsample = [( N/N)sample/( N/N)reference – 1] × 1000 (9) δOsample = [( O/O)sample/( O/O)reference – 1] × 1000 (10) where the 15N/14N reference is N2 in air and the 18O/16O reference is Vienna standard mean ocean water (VSMOW). Referencing to air and VSMOW was through comparison to the international potassium nitrate reference material IAEA-N3, with an assigned δ15N of +4.7‰ (Gonfiantini et al. 1995) and reported δ18O of +22.7 to +25.6‰ (Böhlke et al. 1997; Böhlke et al. 2003; Revesz et al. 1997; Silva et al. 2000). We adopted a δ18O of 25.6‰ (Böhlke et al. 2003), but without consequence, since only isotope ratio differences are reported in this study. Unless indicated otherwise, the N and O isotopic ratios represent the mean of any replicate measurements; measurements of roughly 10% of the samples were duplicated. N and O isotope ratios of nitrate measured for the nitrate consumption experiment with P. aureofaciens were fitted to the Rayleigh isotope fractionation model to determine the isotope Granger et al. Nitrite removal