Fred J. Genthner*, Dragoslav T. Marcovich and John C. Lehrter
US Environmental Protection Agency, Office of Research and Development, National Health and Environmental Effects Research Laboratory, Gulf Ecology Division, Gulf Breeze, FL, 32561, USA
Received date: September 17, 2013; Accepted date: October 30, 2013; Published date: November 04, 2013
Citation: Genthner FJ, Marcovich DT, Lehrter JC (2013) Estimating Rates of Denitrification Enzyme Activity in Wetland Soils with Direct Simultaneous Quantification of Nitrogen and Nitrous Oxide by Membrane Inlet Mass Spectrometry. J Microb Biochem Technol 5:095-101. doi:10.4172/1948-5948.1000108
Copyright: © 2013 Genthner FJ, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Microbial & Biochemical Technology
The microbial mediated process of denitrification is a major pathway for the removal of reactive nitrogen a pollutant in natural environments and waste treatment facilities. Denitrification potential, measured as denitrification enzyme activity (DEA), was quantified in novel short-term (4 h) anaerobic assays using a more sensitive and precise technique, membrane inlet mass spectrometry (MIMS) rather than the traditional headspace electron capture gas chromatography (GC-ECD) method. Using MIMS modifications made to the instrument and sample handling allowed for the simultaneous and direct measurement of reaction products nitrous oxide (N2O), a potent greenhouse gas, and the chemically unreactive dinitrogen (N2). Rate determinations were made from the slope of a linear curve generated by plotting increasing concentrations of the reaction products with time. Strong evidence for the validity of MIMS measured DEA rates was provided by showing consistent, linear accumulations of N2O or N2 and close agreement in rates from replicate reactions. Reactions were performed using wetland soils and cultures of Pseudomonas aeruginosa and P. chloroaphis that generated denitrification end products of N2 and N2O, respectively. Under acetylene inhibition P. aeruginosa produced the N2O end product at a rate equivalent to the rate obtained in the uninhibited reaction that produced N2. No significant (p>0.05) difference was observed between MIMS or headspace with GC-ECD, determined DEA in wetland soil reactions under acetylene inhibition. Because of anoxic conditions in the reaction vessels used with MIMS, detectable rates of N2O accumulation were only observed in acetylene blocked reactions or in cultures of P. chloroaphis. This method has potential applications ranging from near realtime wastewater treatment process measurements to field studies of nitrogen cycling. The continued development and application of these types of methods are needed to improve our understanding of the mechanisms regulating denitrification and its benign, N2, and harmful, N2O, end-products.
Denitrification Enzyme Activity; Membrane Inlet Mass Spectrometry; nitrous oxide
The denitrification enzyme activity (DEA) assay provides an assessment of the multi-enzyme, process of reactive nitrogen removal by the reduction of nitrate to nitrous oxide (N2O) and/or nitrogen gas (N2). Measured in soil slurries under non-limiting carbon and nitrate concentrations, this short term (≤ 4 h) anaerobic assay, also defined as denitrification potential , yields quantitative measures on potential denitrification rate. Denitrification enzyme activity is related to the magnitude of denitrifying enzymes and yields an index of the denitrifier population in soils reflective of long-term variation in the factors that control denitrification . Denitrification capacity (DC) is defined in this manuscript as the endogenous denitrification rate measured in the absence of added nutrients. Chloramphenicol is added to prevent de novo synthesis of denitrification enzymes so that DEA is reflective of the amount of denitrification enzymes at the time of sampling. Denitrification enzyme activity is commonly measured by the production of N2O, which accumulates due to acetylene (C2H2) inhibition of the enzyme nitrous oxide reductase. Nitrous oxide, found in low concentration in the atmosphere, is readily measured by gas chromatography . In the absence of an acetylene block the small amount of the major end product, N2, is difficult to measure in the headspace due to the high background concentration of N2 in the atmosphere .
In water, however, the background concentration of N2 is much lower allowing denitrification to be determined by measuring the accumulation of N2 in aqueous samples. Kana et al.  modified a membrane inlet mass spectrometer to perform precise measurements of dissolved N2 based on the N2/Ar ratio. This analytical capability, plus problems encountered with the C2H2 method limiting its usefulness , prompted Khalil and Richards  to develop a method to measure DEA in soil slurries using membrane inlet mass spectrometry (MIMS). These investigators demonstrated DEA by observing decreased N2 / Ar ratios with increasing soil depth and peaks for the N2/Ar ratios when the soil slurries were amended with optimal concentrations of NO3-N and glucose-C for the expression of DEA.
Denitrification is a miserable process to measure as most methods are problematic for a variety of reasons . The MIMS technique gained popularity for denitrification measurements as it is capable of rapidly (< 2 min) measuring small (< 7 ml) samples with high precision . The purpose of this study was to expand the usefulness of MIMSbased techniques by developing a method that provided a quantitative measure of DEA plus provide the instrument with a greatly needed capability, namely, the ability to directly measure N2O. Validation of these techniques was provided through the use of bacterial controls and by comparing our MIMS based approach to the established, acetylene inhibition, headspace GC-EDC method .
Driven largely by increases in emissions from agricultural activities N2O concentrations in the atmosphere are projected to rise . The application of nitrogenous fertilizers promotes release of N2O via denitrification and nitrification . Although many chemical and physical factors increase the proportion of N2O to N2 released during denitrification , little is known about factors controlling N2O formation by denitrification in sediments of aqueous systems . Therefore, detecting and quantifying N2O by MIMS should increase the efficacy of this analytical technique for studying ways to minimize the release of this potent greenhouse gas in the atmosphere.
Presented in our study are changes made to enhance the efficacy of MIMS to measure denitrification and modifications to the protocol of Khalil and Richards  needed to quantify DEA. These were reaction vessel design, sample treatment, rate calculations, controls and data quality objectives. Reactions were performed and rates calculated in the presence (DEA) and absence (DC) of added glucose and nitrate. Observing higher DEA than DC in the same soil served as a data quality objective. Specific objectives were to: (i) modify the procedure and instrument, mass spectrophotometer, to allow direct simultaneous measurements of N2 and N2O; (ii) develop, using MIMS, a short term (4 h) assay to quantify DEA in wetland soils based on measuring the linear accumulation of N2 and N2O over time; (iii) express MIMSmeasured DEA as a rate, μg N2 or N2O h-1 kg dry weight (wt) soil-1; (iv) compare DEA in a wetland soil under acetylene inhibition using MIMS and headspace with GC-ECD to measure the N2O product; (v) assay pure cultures of denitrifying bacteria as controls; and (vi) examine storage stability of dissolved N2 in samples taken for MIMS analysis.
Wetland sites, soil sampling and soil characteristics
Wetland soils were obtained from three locations in the Pensacola Bay watershed identified, using the U.S. Fish and Wildlife Service National Wetlands Inventory  as freshwater forested/shrub (PF01/4B). Site names and their locations (latitude/longitude) in this northern Gulf of Mexico area are listed in Table 1. Soil samples, collected during the summer of 2012, were obtained from the top 10 cm of soil, passed through a 2-mm sieve and stored at room temperature until analysis. Denitrification enzyme activity or DC was measured in soil samples within 18 h of collection. A portion of the sample, 30 to 50 g, was weighed before and after drying at 110°C to calculate dry weight (dry wt). The percent organic matter was then calculated by the loss in weight after combustion for 18 h at 440°C .
|Site||Latitude/Longitude||Organic Carbon (%)||Denitrification capacity, no added nutrientsa||DEA, added glucose and nitrateb|
|µg N2 h-1 kg dry wt soil-1 ± Std Devc|
|Nature Reserve||30° 38.335' N/87° 04.525’ W||9||4.4 ± 0.3||133 ± 13|
|Shoreline Park South||30° 21.109' N/87° 10.554' W||20||114 ± 86||525 ± 58|
|Lake Shore Trail||30° 21.127' N/87° 10.096' W||41||UDd||576 ± 98|
aDenitrification reactions with only endogenous levels of nitrate and glucose. bNutrients for denitrification were made non-limiting by the addition of nitrate and glucose to the reaction vessels. cRates are expressed as the mean of triplicate determinations. Based on the results of t-tests (p>0.05), the null hypotheses were rejected. Thus, within each soil the denitrification enzyme activity rates were higher than the denitrification capacity rates. dUD, unable to determine (no increase in the N2 /Ar ratio was observed)
Table 1: Denitrification enzyme activity rates determined in short term (4 h) anaerobic assays in the presence and absence of nutrient (NO3-N, glucose-C) amendments. The nitrogen product was measured by membrane inlet mass spectrometry.
Bacteria and growth conditions
Denitrification end products were also studied from pure cultures of two denitrifying bacteria, Psuedomonas aeruginosa (ATCC 15692) and P. chloroaphis (ATCC 9446), where they served as controls to compare to the soils samples since the former produces only N2  and the latter produces only N2O (15). Bacteria were cultured at 25°C for 48 h without shaking in nitrate broth (Difco, Detroit, MI) contained in completely filled 100 ml screw-caped media bottles (Corning, Corning, NY). Dry weight of cells was determined by filtering 20 ml of culture through pre-weighed 0.40 μm polycarbonate filters. Filters were dried in a desiccator (25°C) to a constant weight and reweighted to calculate dry weight of cells per unit volume.
Reaction conditions when MIMS was used to measure end products of DEA or DC rates
The reaction vessel was a 160 ml glass, serum stoppered bottle equipped with two ports. One port was for removal of sample from the top, and the other to add fresh DEA or DC solution at the bottom of the vessel to restore volume after sample removal. Ports were provided with 8 gauge, BD (Franklin Lakes, NJ) Luer-LocTM needles inserted through the septum. A one-way valve was attached to the Luer-Loc connector. The distal end of the 1-way valve was fitted with either a 22 cm long tube (Tygon®, I.D. 5 mm) for the sample removal or a 10 ml syringe for DEA or DC addition (Figure 1A).
Approximately 50 g wet wt of soil or 20 ml of a bacterial culture was added to the reaction vessels followed by a sufficient volume of helium (He) flushed DEA or DC solution to enable the vessels to be sealed, without air entrapment, with a rubber stopper secured by a crimped aluminum cap. When used to block the reduction of N2O to N2, 10 ml of atomic absorption grade, 99.6% purity, acetylene (C2H2), was added to each vessel using a gas tight syringe. The acetylene was forced to slowly dissolve into the liquid by applying constant pressure to the plunger. The vessels were inverted several times. Two ports were inserted through the rubber septum. The remaining bubbles were removed through one port by displacement with DEA or DC solution added through a second, replenishment port. DEA solution, which made conditions for denitrification non-limiting, contained the following: KNO3, 1.4 mg ml-1; glucose, 1 mg ml-1 and chloramphenicol, 0.025 mg ml-1. The DC solution lacked both nutrients. Reaction vessels were incubated with shaking at 25°C. At approximately 0.5, 1, 2, 3 and 4 h samples were removed via displacement of sample by injecting 9 ml of DEA or DC through the replenishment port. After each sample was removed the vessel was inverted to assure sufficient mixing. Samples were collected into 8 ml ground glass stoppered test tubes containing 50 μl of a saturated solution of sodium hydroxide stopping the reaction and converting CO2, which interferes with quantification of N2O, into sodium bicarbonate. The completely filled sample tubes were placed under water in test tube racks held in an appropriate size ice chest. Samples were stored at room temperature until analysis.
Analysis of dissolved N2, N2O, Ar and O2 by MIMS
Dissolved gases were measured on a Bay Instruments dissolved gas analyzer (Easton, MD) that used Pfeiffer Vaccuum’s Prisma mass spectrometer (Aslar, Germany). A "Shuttle Portable" ultralow freezer, Stirling Ultracold (Athens, OH) provided a -86°C cold trap (Figure 1B). The gas ratio method, employing argon (Ar) as the tracer , was used to calculate μg of dissolved N2, N2O, and O2 in the reaction vessels. Calibration factors were used to adjust the N2/Ar, O2 /Ar and CO2/Ar ratios measured in air-equilibrated deionized water (dH2O, 20°C) to the ratios reported by Colt . Specifically, these ratios were 38.187 for N2/Ar, 20.393 for O2/Ar, and 0.879 for CO2/Ar . The reported CO2/Ar ratio served as a surrogate to determine a N2O/Ar correction factor for calibrating the instrument. Carbon dioxide was an appropriate surrogate as both CO2 and N2O possess a molecular mass of 44, and the solubility of CO2 is similar to that of N2O . Furthermore, a surrogate was necessary as the concentration of N2O in air saturated water is orders of magnitude less than CO2 . To calculate the mass (μg) of each gas in the reaction vessel at the time of sampling the corrected ratios were multiplied by μM of Ar in dH2O at air saturation (13.91 μM, 16). These μM concentrations were multiplied by their respective molecular weights and the volume of the reaction vessel (160 ml) to yield μg of each gas. A correction factor was used to account for loss of product from sampling.
Reaction conditions when GC-ECD was used to measure DEA rates by acetylene inhibition (headspace) method
DEA was measured as the mean of triplicate determinations using a modified version of the headspace with GC-EDC method described by White and Reddy  in which denitrification conditions are made non-limiting by the addition of excess nitrate and glucose. To each 160 ml glass reaction vessel was added approximately 50 g wet wt soil and 50 ml DEA solution. The slurry was made anaerobic by flushing with N2 gas prior to crimp sealing the vessel. Ten ml of headspace gas was replaced with 12 ml of C2H2 (approximately 15% of the headspace). The vessels were incubated with shaking at 25°C. Gas samples, removed at approximately 0.5, 1, 2, 3 and 4 h for N2O analysis, were immediately injected into a Hewlett-Packard (Palo Alto, CA) 6890 GC with ECD adapted for direct gas injection through an 8-port valve with a 0.25 μL sample loop. Separations were performed using an Agilent (Santa Clara, CA) GS-Q megabore column (30 m X 0.53 mm I.D., porous divinyl homopolymer/durabond) run at 40°C.
Stability of dissolved nitrogen in samples stored for MIMS analysis
To explore the time constrains of performing DEA reactions at a laboratory lacking a membrane inlet mass spectrometer and transporting the samples to a different facility for MIMS analysis, the loss of dissolved N2 in stored samples was measured over time. A denitrification reaction was set up as follows: soil, 325 g wet wt obtained from the Lake Shore Trail site, was placed in a 1 liter screw cap bottle and filled with DEA solution without air entrapment. Fortyfive samples were removed from the reaction vessel after 8 h at 25°C, placed into sample storage tubes and held under water as described previously. After 1, 3, 6, 8 and 10 days nine replicate samples were removed and analyzed by MIMS for loss of dissolved nitrogen.
Statistical analyses were performed using SAS for Windows, release 9.2 (SAS Institute, Inc., Cary NC, USA). A two-sample t-test was used to compare the means, at the 95% confidence interval, of DEA and DC activities measured in soils or pure cultures of bacteria. A single factor ANOVA combined with a Tukey’s Studentized Range (HSD) test was used to determine the storage time at which the means of the nitrogen concentrations in the samples were significantly (p<0.05) lower than the initial concentration.
Modifications to allow the direct measurement of nitrous oxide by MIMS
The membrane inlet mass spectrophotometer was reconfigured and samples alkalinized to directly measure N2O in addition to N2, Ar and O2. Overlying waters , heavily polluted water  and wastewater treatment plants  are all aquatic environments where simultaneous fluxes of these gases have been shown to occur. Furthermore, using MIMS to measure denitrification in flow-through systems by only the N2/Ar ratio may underestimate this reaction if reduction stopped at nitrous oxide rather than continuing to completion .
In the current study a mixture of the two end products in the reactions was not observed. This finding was in agreement with the results of Khalil and Richards  who, using GC-ECD to measure N2O gas that had diffused into a helium headspace from aqueous samples taken from their DEA reactions, found that net N2O production was remarkably lower than N2 production and could be ignored for estimating DEA or DC.
To allow the direct measurement of N2O by MIMS, modifications to the instrument and procedure were necessary. Both carbon dioxide (CO2) and N2O possess the same atomic mass and, therefore, N2O cannot be assayed directly at their mass/charge ratio (m/z) of 44. Previously, N2O was measured in the presence of CO2 by determining the cracking patterns produced by the mass spectrometer and assaying m/z of 28, 30, and 32 from solutions of defined dissolved gas concentrations . A similar approach was used by Firth and Edwards  who distinguished between CO2 and N2O in denitrifying cultures of Pseudomonas stutzeri by studying minor ion peaks created from these gasses. However, in the current study, a simpler and more direct measure of N2O was achieved by removing the CO2 from the sample. This was accomplished by adding NaOH to the sample to convert the dissolved CO2 to bicarbonate. The instrument was also equipped with a cold trap maintained at -86°C. This temperature is below the freezing point of CO2 (-78°C) yet above the freezing point of N2O (-91°C). If a small ultralow freezer is unavailable for use as a cold trap, a mixture of liquid N2 and ethyl methyl ketone will attain the same temperature .
Validation of MIMS-based protocol for measuring DEA and DC rates in soils
In the following three sections we explain how the MIMS-based protocol for measuring DEA and DC rates was validated namely by: i) observing a linear accumulation of the N2O or N2 product; ii) measuring higher DEA than DC rates in soils; iii) assaying two control bacterial cultures that differed in their denitrification end products; and iv) obtaining comparable DEA rates in a wetland soil measured under acetylene inhibition using our MIMS-based and the headspace GC-ECD methods.
Denitrification enzyme activity and DC rates determined in soils using MIMS to measure the N2 product
The name of each study site, its geographic coordinates and % organic matter in its soils are presented in Table 1; also shown are MIMS-determined DC and DEA rates. A linear accumulation of the N2O or N2 product was observed. An R2 value ≥0.85 for all linear rates was achieved. Standard deviations of triplicate DEA determinations for all wetland soils were <20% of the mean. The Shoreline Park soil showed high variability in its DC. Perhaps non-limiting nutrients conditions in DEA reactions buffered the effects of elevated concentrations of NO3 - and/or carbon present in very small activity centers in Shoreline Park soil . Within each site DEA activities were significantly higher than DC activities. Results of a two-sample t-test performed between means of DC and DEA activities for Shoreline Park soils provided evidence to reject the null hypothesis (p=0.002, t-value=6.85, df=4). Likewise results of a two-sample t-test performed between means of DC and DEA activities for the nature reserve soils also provided evidence to reject the null hypothesis (p=0.0005, t-value=44, df=2). These findings not only contributed to validating the method but also suggested nutrient limitation(s) (Table 1). A nitrogen limitation was probably affecting DC activities in all soils assayed as it is frequently limited in wetlands . It is unlikely that carbon was limited in histosols (20% or greater organic matter by weight) of the Shoreline Park and Lake Shore Trail sites. The strongest indication of nitrogen limitation was in Lake Shore Trail where the DEA activity was very high yet the DC activity was below detection (Table 1).
Use of bacterial cultures for validating the MIMS-based DEA method
Pure cultures of bacteria provided control and validated our MIMS-based methods (Table 2, Figure 2). The dynamics of a typical P. aeruginosa DEA reaction in the presence and absence of C2H2 was illustrated by plotting changes in N2 /Ar, N2O/Ar and O2/Ar over time (Figure 2A,B). In both reactions the accumulation of either denitrification end product was accompanied by a concomitant decrease in the O2/Ar ratios over the 4 h reaction period. This decrease in oxygen, the preferred electron acceptor, was most rapid during the first hour of the reaction (Figure 2A, B). Under these highly reduced conditions in the absence of C2H2, a complete reduction of NO2 to N2 and no increase in the N2O/Ar ratios was observed (Figure 2A, Table 2). In this reaction a linear increase in the N2 /Ar ratios (R2 = 0.98) was observed.
|Bacterial culture||Acetylene added||Rate|
|µg N2 h-1 mg dry wt cells-1||µg N2O h-1 mg dry wt cells-1|
|Pseudomonas aeruginosa||No||22.6 ± 6.8a,b||35.5 ± 10.7 a,b,c|
|Pseudomonas aeruginosa||Yes||UDd||25.8 ± 3.3a,b|
|Pseudomonas chloroaphis||No||UDd||2.9 ± 1.0b,e|
|Pseudomonas chloroaphis||Yes||UDd||2.3 ± 1.1b,e|
amean of five determinations; bBased on the results of t-tests (p > 0.05), we failed to reject the null hypothesis that the means of the rates obtained the presence and absence of acetylene with either P. aeruginosa (single underline) and P. chloroaphis (double underline) were the same; cValue was not measured directly as N2O was not produced. The rate was obtained by converting the rate, measured by the production of N2, to N2O h-1 mg dry wt cells-1; dUD, unable to determine as N2 was not produced. emean of triplicate determination
Table 2: Denitrification enzyme activity rate (± Std Dev) of bacterial cultures determined by measuring accumulation of the nitrogen or nitrous oxide end product by membrane inlet mass spectrometry.
In the presence of C2H2, which blocks nitrous oxide reductase , P. aeruginosa produced only N2O as the sole end product (Figure 2B). Thus, DEA was measured as an increase in N2O over time (Table 2). However, in the absence of C2H2 in the P. aeruginosa culture DEA was measured as an increase in N2 over time. To use a two sample t-test for determining whether the mean DEAs were significantly different in the presence or absence of C2H2, these rates needed to be expressed in the same units. Therefore, DEA, measured in the absence of C2H2, was also expressed in terms of N2O produced (Table 2). This was accomplished by multiplying the measured N2 rate by 1.57 (44 MW of N2O/28 MW of N2). Results of an unequal variance t-test (p=0.38, t-value=0.96, df=6) provided evidence to accept the null hypothesis of equal means.
Pseudomonas chloroaphis (ATCC 9446), lacking nitrous oxide reductase, only produced N2O as the gaseous end product of denitrification (Table 2). Therefore, a DEA rate could only be directly measured as the accumulation of N2O over time. These reactions were run in the presence and absence of C2H2 and mean DEA rates were determined. Results of an equal variance t-test (p=0.51, t-value=0.72, df=4) provided evidence to accept the null hypothesis of equal means. This indicated that C2H2 did not affect the rate of denitrification. Significantly higher mean DEA rates were measured in P. aeruginosa than in P. chloroaphis cultures (Table 2).
Comparison of DEAs in soil determined under acetylene inhibition using the MIMS-based method versus the headspace with GC-ECD method to measure the N2O product
DEA rates were measured in soil from the Shoreline Park site by the, widely used, acetylene inhibition (headspace) method that employs GC-EDC to measure the N2O product and the MIMS-based method developed in the present study. Acetylene was used with the MIMSbased method to allow direct comparisons with a common end product, N2O. Mean DEA rates of 853 ± 302 μg N2O h-1 Kg dry wt soil-1 and 896 ± 431 μg N2O h-1 Kg dry wt soil-1 were measured with the MIMS-based and headspace with GC-ECD based methods, respectively. DEAs were expressed as the mean of triplicate determinations. Within methods, the headspace with GC-ECD method showed the highest variation; the standard deviation was 48% of the mean rate; whereas with the MIMS-based method the standard deviation was only 35% of the mean rate. Perhaps different reaction vessel conditions, one completely filled with soil slurry and the other possessing a headspace, was partially responsible for the differences in variability. A significant difference was not observed between means of the rates measured by either method.
Sensitivity of measuring N2O by MIMS
An analysis of the sensitivity of N2O measurement by MIMS was warranted as it may be of interest when applying this measurement to low flux systems. As the concentration of N2O in air saturated water is extremely low and reliable standards are unavailable, instrument noise was used to gauge sensitivity with the premise that N2O/Ar ratios higher than instrument noise could provide a sensitivity assessment. Instrument noise was calculated as the mean (± Std Dev) of 15 N2O/ Ar ratios measured in alkalized dH2O (200 C, pH 14) after instrument stabilization at approximately 4 minute intervals over a one hour time period. Three standard deviations above the mean N2O/Ar was selected as the level of sensitivity as there is only a 0.3 % probability that normally distributed ratios lie above this value . A mean N2O/Ar ratio (± Std Dev) of 0.097 ± 0.011 was measured as instrument noise. An instrument calibration factor of 1.97 for the N2O/Ar ratio was obtained as described in the Materials and Methods section. Multiplying by this factor yielded a corrected mean N2O/Ar ratio (± Std Dev) of 0.192 ± 0.022. At three standard deviations above this mean, a ratio of 0.258 or greater would have at least a 99.7% probability of representing a true increase in N2O above instrument noise. Multiplying this ratio by the concentration of Ar in air saturated dH2O at 200 C (13.91 μM, 16) suggested that a minimum N2O concentration of 3.59 μM could be measured with the aforementioned degree of confidence. Furthermore, three times the standard deviation of the corrected N2O/Ar ratio is 0.066. This value again multiplied by the concentration of Ar in air saturated dH2O at 200 C (13.91 μM, 16) suggested that a 0.92 μM increase in the dissolved N2O concentration of above 3.59 μM could be measured with the same degree of confidence.
Stability of dissolved N2 in samples stored for MIMS analysis
Work at laboratories not possessing a membrane inlet mass spectrometer may require sample transport and/or storage before analysis. To explore the time constrains of performing DEA reactions at a laboratory lacking a membrane inlet mass spectrometer and transporting the samples to a different facility for MIMS analysis, the loss of dissolved N2 in stored samples was measured over time. Samples held for MIMS analysis were examined for loss of dissolved N2 after 1, 3, 6, 8 and 10 days of storage (Figure 3). Replicate samples, supersaturated with N2, were removed from a denitrification reaction and stored underwater in ground glass-stopper test tubes as described previously. At each time period nine samples were analyzed for loss of N2 by leakage from the storage tubes. No significant loss of N2 was measured in the sample storage tubes after eight days. On day 10 the mean N2 concentration in the storage tubes decreased. A single factor ANOVA showed that this average N2 concentration was significantly lower than the average N2 concentration in the storage tubes held for all shorter time periods. Results demonstrated that samples could be stored under water in the ground glass stopper collection tubes for a sufficient length of time before analysis to allow flexibility for MIMS- based denitrification studies to be conducted at remote locations (Figure 3).
The membrane inlet mass spectrometer was reconfigured and samples alkalized to allow the direct measurement of N2O. Using MIMS DEA and DC were measured short term assays as rates. These protocols were validated by i) observing a linear increase in end products; ii) observing close agreement between replicates; iii) measuring higher DEA than DC capacity in the same soil; iv) obtaining comparable DEA rates in a soil measured under acetylene inhibition using our MIMS-based and the headspace with GC-ECD method; v) using pure cultures of bacteria that differed in their denitrification end products. Furthermore, an approach to assess the sensitivity of measuring N2O by our MIMS-based method in the absence of reliable standards suggested that a minimum N2O concentration of 3.59 μM could be measured with a 99.7% probability of confidence. Finally, to determine the time constrain imposed by transporting samples to a different facility for MIMS analysis revealed that samples could be stored submerged in water for 8 days without a significant loss of N2.
We wish to thank Marc Russell for assistance with the statistical analyses. Thanks also to Matthew Harwell, Ken Forshay and Todd Kana for useful comments on the manuscript. The information in this document has been funded wholly (or in part) by the U.S. Environmental Protection Agency. It has been subjected to review by the National Health and Environmental Effects Research Laboratory and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.