Article
A novel method for nitrogen isotopic analysis of soil-emitted nitric oxide (NO) Zhongjie Yu, and Emily M. Elliott Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 03 May 2017 Downloaded from http://pubs.acs.org on May 8, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 44
Environmental Science & Technology
1
A novel method for nitrogen isotopic analysis of soil-
2
emitted nitric oxide (NO)
3
Zhongjie Yu* and Emily M. Elliott
4
University of Pittsburgh, Department of Geology and Environmental Science, Pittsburgh,
5
PA 15260, USA
6
* Corresponding author. Phone: (973) 718 0238; Fax: (412) 624 8780; Email:
[email protected] 7 8 9 10 11 12 13 14 15 16 17 18 19 20
ACS Paragon Plus Environment
Environmental Science & Technology
21
ABSTRACT
22
The global inventory of NOx (NOx=NO+NO2) emissions is poorly constrained with a
23
large portion of the uncertainty attributed to soil NO emissions that result from soil
24
abiotic and microbial processes. While natural abundance stable N isotopes (δ15N) in
25
various soil N-containing compounds have proven to be a robust tracer of soil N cycling,
26
soil δ15N-NO is rarely quantified due to the measurement difficulties. Here, we present a
27
new method that collects soil-emitted NO through NO conversion to NO2 in excess ozone
28
(O3) and subsequent NO2 collection in a 20% triethanolamine (TEA) solution as nitrite
29
and nitrate for δ15N analysis using the denitrifier method. The precision and accuracy of
30
the method were quantified through repeated collection of an analytical NO tank and
31
inter-calibration with a modified EPA NOx collection method. The results show that the
32
efficiency of NO conversion to NO2 and subsequent NO2 collection in the TEA solution
33
is >98% under a variety of controlled conditions. The method precision (1σ) and
34
accuracy across the entire analytical procedure are ±1.1‰. We report the first analyses of
35
soil δ15N-NO (-59.8‰ to -23.4‰) from wetting-induced NO pulses at both laboratory
36
and field scales that have important implications for understanding soil NO dynamics.
37 38
TOC Art
39
ACS Paragon Plus Environment
Page 2 of 44
Page 3 of 44
Environmental Science & Technology
40
INTRODUCTION
41
Emissions of nitrogen oxides (NOx = NO + NO2) degrade air quality and affect global
42
tropospheric chemistry,1, 2 posing a significant danger to ecosystem and human health.3-6
43
Although fossil fuel combustion is currently the largest source of atmospheric NOx,7, 8
44
NO is also produced in and emitted from natural and fertilized soils.9-11 Due to the spatial
45
segregation of different NOx sources and the short boundary layer lifetime of NOx, there
46
are substantial areas of the world (e.g., tropical and agricultural regions) where the local
47
NOx budget is controlled exclusively by soil NO emissions.3, 7, 12-15 In these regions, soil
48
NO emissions govern the formation and lifetime of tropospheric ozone (O3) and hydroxyl
49
radical, driving reaction chains that produce environmentally important trace gases (e.g.,
50
nitric acid and peroxyacetyl nitrate) and biogenic secondary aerosols.4, 13, 16
51
Various processes, both microbial10, 17-19 and abiotic20-22, are capable of producing
52
NO in soils. Although the strong dependence of soil NO emission on edaphic and
53
climatic factors has long been demonstrated by laboratory and field studies,23-27 a
54
process-based understanding of soil NO dynamics is lacking.14, 28 More importantly, soil
55
NO emission often exhibits an episodic nature (e.g., time scale of minutes), with pulse-
56
like emission events being often triggered by rewetting of dry soils.12, 13, 29-33 In dry
57
agricultural soils, massive NO pulses triggered by coupled fertilization and precipitation
58
during warm seasons can result in daily O3 enhancement up to 16 ppbv.13 Unfortunately,
59
the sources of and processes controlling the pulsed soil NO emission are still
60
mysterious,13, 14 making it difficult to model and up-scale field-observed NO fluxes.
61
While empirical bottom-up models estimate that soil NO emission accounts for about
62
15% of the global NOx inventory,1 inversion of satellite-based NO2 observations have
ACS Paragon Plus Environment
Environmental Science & Technology
63
indicated significant underestimates in soil NO emission (e.g., up to a factor of 3) at
64
various spatiotemporal scales.7, 12-16, 29, 34 Indeed, with the substantial reductions in NOx
65
emissions from combustion sources in many countries,2, 35 soils as a source of
66
atmospheric NOx may be more important than we thought, and there is a pressing need to
67
elucidate mechanisms underlying soil NO dynamics.13, 14
68
Stable nitrogen (N) isotope compositions at natural abundances (notated as δ15N)
69
in various soil N-containing compounds are a robust tracer of soil N cycling.36-39
70
Incorporation of δ15N-NO measurements into the soil N isotope systematics is expected
71
to provide a new process-level information of key mechanisms regulating NO production
72
and consumption in soil. Moreover, there is a growing interest in using δ15N of
73
atmospheric N oxides (e.g., NO2 and nitrate (NO3-)) as a tracer to partition NOx emission
74
sources over large spatial and temporal scales.40-43 This interest stems from the
75
observations that NOx emitted from different sources has distinct δ15N values44-49 and that
76
soil-emitted NO is presumably lower in δ15N than NOx from other natural and
77
anthropogenic sources.50-52
78
Despite its promising potential, soil δ15N-NO is rarely measured due to the
79
intermittent nature and low magnitudes of soil NO emission. A summary of published
80
NOx collection methods is provided in Table S1 in the supporting information (SI)),
81
highlighting that none of existing methods have been rigorously verified for their
82
suitability for soil-emitted NO. In pioneering work, Li and Wang50 fertilized a soil
83
monolith in the laboratory and collected NO by first converting NO to NO2 using a
84
chromium trioxide (CrO3)-impregnated solid oxidizer and then trapping the converted
85
NO2 in an annular denuder as nitrite (NO2-) for δ15N analysis. However, it is well
ACS Paragon Plus Environment
Page 4 of 44
Page 5 of 44
Environmental Science & Technology
86
documented that NO oxidation efficiency of the CrO3 oxidizer varies dramatically with
87
sample relative humidity (RH) (e.g., 60%).53, 54 Due to this overlooked
88
humidity interference, it is unclear whether N isotopic fractionation can occur during the
89
NO oxidation under varying soil conditions. Recently, Fibiger et al.55 and Wojtal et al.56
90
presented a NOx collection method that utilizes a KMnO4+NaOH solution to actively
91
collect NOx as NO3- for δ15N-NOx determination. Sample δ15N-NOx must be calculated
92
using an isotope mass balance due to a high reagent blank in the solution (5-7 µM NO3-,
93
δ15N=~2‰)49, 55. While the precision of this approach is ±1.5‰, this method is
94
incompatible for δ15N-NO measurement of low and diffuse soil NO emissions, because
95
larger error is propagated from the isotope mass balance calculation if concentration and
96
δ15N value of collected soil NO are significantly lower than the blank.
97
Here, we present a new method for soil δ15N-NO determination (hereafter, “DFC-
98
TEA method”). This method collects NO through NO conversion to NO2 in excess O3
99
and subsequent NO2 collection in a triethanolamine (TEA) solution as NO2- and NO3- for
100
δ15N analysis. The NO collection approach is coupled to a soil dynamic flux chamber
101
(DFC) system for simultaneous NO flux and δ15N-NO measurements. Both laboratory
102
and field method verifications have been conducted to demonstrate suitability of the
103
DFC-TEA method for accurate and precise soil δ15N-NO determination.
104 105
EXPERIMENTAL SECTION
106
DFC system setup
107
The DFC is a technique that has been developed to continuously measure soil-atmosphere
108
fluxes of various compounds including NO.18, 19 A schematic of the developed DFC
ACS Paragon Plus Environment
Environmental Science & Technology
109
system is shown in Figure 1. The system consists of five components: air purification unit,
110
gas dilution unit, flux chamber, NO-NOx-NH3 analyzer, and NO collection train. Zero air
111
free of NOx and O3 is produced in the air purification unit for purging the flux chamber
112
and providing air to a O3 generator (Model 146i, Thermo Fisher Scientific) in the NO
113
collection train. NO, NO2, and ammonia (NH3) concentrations in the chamber headspace
114
are measured alternately by a chemiluminescent analyzer (Model 17i, Thermo Fisher
115
Scientific) at 10 s intervals for flux calculations. For method development, reference NO,
116
NO2, and NH3 from three analytical tanks were diluted into the purging flow to simulate
117
soil gas emissions inside the chamber. Two versions of the DFC system were developed
118
for laboratory and field experiments. In the laboratory DFC system, a 1 L Teflon flow-
119
through jar is used as the flux chamber. For the field DFC system, we fabricated a
120
cylindrical flow-through chamber (39 cm I.D. and 30 L inner volume; Figure 1b),
121
following considerations57, 58 for minimizing pressure differentials in chamber headspace.
122
Control tests indicate that NO transmission from the chamber is greater than 98.3%.
123
Details about the flux measurement, the chamber tests and the specifications for each
124
DFC component are provided in the SI.
125
NO collection train
126
To collect NO for δ15N-NO analysis, a Teflon-coated diaphragm pump is used to sample
127
chamber air passing through the NO collection train (Figure 1). The sample flow rate (1.6
128
standard liter per minute (slpm)) is controlled by a mass flow controller. For the NO
129
conversion in excess O3, a length of Teflon tubing (9.5 mm I.D., ca. 240 cm length)
130
serves as the reaction tube. A O3 flow of 0.4 slpm, produced from photolysis of O2 in
131
zero air at 185 nm by the O3 generator, is mixed with the sample flow at the starting point
ACS Paragon Plus Environment
Page 6 of 44
Page 7 of 44
Environmental Science & Technology
132
of the reaction tube (Figure 1). To prevent generation of HOx radicals during the
133
photolysis, water vapor is removed from the zero air using two drying columns and a
134
Teflon filter is attached before the O3 addition point to decompose remaining HOx
135
radicals.59 Long-term (5 months) average O3 concentration after the mixing of the sample
136
and O3 flows was 2911±32 ppbv as measured by an O3 monitor (Model 202, 2B
137
Technologies). The flow leaving the reaction tube is forced to pass through a 500 mL gas
138
washing bottle with a fritted cylinder containing a solution of TEA (Fisher Scientific,
139
Certified Grade) in water (20% (v/v), 70 mL). The stopper of the gas washing bottle was
140
lengthened so that 70 mL of the solution just covered the frit.
141
Determination of reaction time
142
Reaction of NO with excess O3 forms NO2 (R1 in Table S3 in the SI). In a dark
143
environment, the efficiency of NO to NO2 conversion is limited by the formation of
144
higher nitrogen oxide species (i.e. nitrate radical (NO3) and dinitrogen pentoxide (N2O5);
145
R2-R5 in Table S3 in the SI).59-63 In order to model the NO conversion in the reaction
146
tube, the reaction time is needed. Following Fuchs et al.,61 the reaction time in the
147
reaction tube was experimentally determined by sampling zero air that contained a
148
constant NO concentration (27 ppbv) using the NO collection train and varying the
149
excess O3 concentrations (266-2890 ppbv). The ending point of the reaction tube was
150
attached to the sampling inlet of the chemiluminescent analyzer for NO concentration
151
determination. The NO concentration decay was then fitted to a single exponential
152
function assuming pseudo-first order loss of NO in excess O3 (details are described in the
153
SI). Due to the inner tubing of the chemiluminescent analyzer, the estimated reaction
154
time essentially includes the reaction tube plus the analyzer inner tubing. To correct this
ACS Paragon Plus Environment
Environmental Science & Technology
155
overestimate, the reaction time of the inner tubing was estimated by repeating the
156
experiment with the mixing point of the sample and O3 flow directly attached to the
157
analyzer inlet for NO concentration determination.
158
Preparation of TEA solution
159
Triethanolamine is a tertiary amine and has long been used to scrub acidic gases in fuel
160
gas treating processes and to coat passive filters for ambient NO2 monitoring.52, 64-66 We
161
used a 20% TEA solution for NO2 collection. Its reagent N blank was determined to be
162
0.12±0.04 µM (details are given in the SI).
163
It is reported that aging of the TEA solution can cause significant efficiency
164
decrease in collecting NO2.55 This aging problem may occur to a greater degree with
165
more diluted TEA solutions.55 Therefore, to minimize alteration of TEA from its original
166
state, we sub-sampled new TEA (i.e., freshly opened bottle) into 15 mL glass vials in a
167
glovebox with a 95% N2 + 5% H2 atmosphere to avoid contact with ambient air. Vials
168
were then capped, tightly wrapped with Parafilm, sealed in Ziploc bags, and stored under
169
dark at 4 °C until further use. One glass vial was opened to make a fresh 20% TEA
170
solution immediately prior to each sample collection. The storage time of TEA used in
171
this study was up to approximately 4 months since sub-sampling.
172
Measurement of NO2- and NO3- in TEA solution
173
Both NO2- and NO3- can be produced from the reaction between NO2 and TEA.64-66 NO2-
174
+NO3- concentration in the TEA collection samples was measured using a modified
175
spongy cadmium method.67 Detailed measurement protocol is given in the SI. Control
176
tests using 10 µM NO2- or NO3- in 20% TEA solution indicate that the precision (1 σ,
177
n=8) of the method is ±0.09 µM and ±0.36 µM for NO2- and NO3- measurements,
ACS Paragon Plus Environment
Page 8 of 44
Page 9 of 44
Environmental Science & Technology
178
respectively. Due to the multiple reduction and neutralization steps during the
179
measurements and the N blank inherent to the 20% TEA solution (~0.12 µM), standards
180
were always prepared in 20% TEA solution for concentration calibration.
181
Isotopic analysis
182
The isotopic composition of collected NO2- and NO3- in the TEA solution was measured
183
using the bacterial denitrifier method.68-69 In brief, denitrifying bacteria lacking the N2O
184
reductase enzyme (Pseudomonas aureofaciens) are used to convert 5-20 nmol of NO2-
185
and NO3- into gaseous N2O. Using He as a carrier gas, the N2O is then purified in a series
186
of chemical traps, cryo-focused, and finally analyzed on a GV Instruments Isoprime
187
Continuous Flow Isotope Ratio Mass Spectrometer at m/z 44, 45, and 46 at the University
188
of Pittsburgh Regional Stable Isotope Lab for Earth and Environmental Science
189
Research.
190
Special considerations were taken during the isotopic analysis to ensure precise
191
and accurate measurement of the δ15N of the TEA collection samples. First, the TEA
192
collection samples were neutralized using 12 N HCl to pH ~8 before sample injection to
193
avoid overwhelming the buffering capacity of the bacterial medium.70 Second, in light of
194
the expected low δ15N of soil-emitted NO50-52 and the presence of NO2- as the dominant
195
collection product (see below), a NO2- isotopic standard with low δ15N value (KNO2,
196
RSIL20, USGS Reston; δ15N = -79.6‰, δ18O = 4.5‰)70 is used together with other
197
international NO3- reference standards (IAEA-N3, USGS34, and USGS35) to calibrate
198
δ15N and δ18O measurements. Third, following the IT principle (i.e., identical treatment
199
of sample and reference material), a blank-matching strategy is used to make the isotopic
200
standards in the same matrix (i.e., 20% TEA) as collection samples and to match both the
ACS Paragon Plus Environment
Environmental Science & Technology
201
molar N amount and injection volume (±5%) between the collection samples and the
202
standards (Figure 2). This ensures that the isotopic interference of any blank N associated
203
with the bacterial medium68, 71 and the TEA solution is minimized. The percentage
204
difference (Pdiff) in the major N2O (m/z 44) peak area between each collection sample and
205
RSIL20 measured within the same batch is calculated to quantify how precisely the
206
blank-matching strategy is implemented (Figure 2). Finally, the ∆17O (∆17O=δ17O -
207
0.52×δ18O)72 of the analyte N2O is independently measured for collected samples with
208
sufficient concentration for 50 nmol injection using the N2O thermal decomposition
209
method.73 The resolved ∆17O is then used to correct the isobaric interference on the δ15N
210
analysis resulting from the NO oxidation by O3 according to Kaiser et al.73
211
Quantification of method precision and accuracy
212
The precision of the DFC-TEA method was quantified through repeated NO collection
213
using the reference NO tank (50.4 ppmv). The collection was conducted under a variety
214
of conditions, including differing NO concentrations (12-749 ppbv), chamber
215
temperatures (11.5-30.8 °C), RH (27.1-92.0%), purging flow rates in the field chamber
216
(5-20 slpm), and coexistence of NH3 in high concentrations (500 ppbv). In light of the
217
high temporal variability of soil NO emission, we limited the collection time for each
218
sample to be less than 2 h. Given that soils can produce and emit nitrous acid (HONO)74,
219
75
220
the δ15N-NO analysis by soil HONO emission was minimized by forcing the sample flow
221
to pass through a HONO scrubber (250 mL fritted gas washing bottle containing 50 mL
222
of 1 mM phosphate buffer solution at pH 7.0)76 before entering the NO collection train.
and that HONO positively interferes NO2 collection in TEA solution,65 interference of
ACS Paragon Plus Environment
Page 10 of 44
Page 11 of 44
223
Environmental Science & Technology
While there is no certified isotopic standard for gaseous NO, the accuracy of the
224
DFC-TEA method was evaluated through inter-calibration with a modified EPA NOx
225
collection method. A detailed description of the modified EPA method has been provided
226
by Felix et al.44 and Walters et al.47 In brief, gas samples from the NO and NO2 tanks
227
were collected directly into evacuated 1 L borosilicate gas sampling bulb containing 10
228
mL of a NO2 absorbing solution (H2SO4+H2O2) on a vacuum line. The absorbing solution
229
oxidizes NO2 into NO3-. For the NO collection, the collection was terminated with a
230
small vacuum remaining in the bottle. The bottle was then quickly vented to the
231
laboratory atmosphere to allow introduction of O2 into the bottle for the conversion of
232
NO to NO2.77 After the collection, the bottles were allowed to stand for 1 week with
233
occasional shaking to facilitate the conversion of NOx to NO3-. The residual NOx
234
headspace concentration was measured after a 1 week period and indicates that the
235
collection was 100%. The absorbing solution was then collected and neutralized for δ15N
236
analysis using the denitrifier method. The results show that the NO and NO2 tanks had
237
δ15N values of -71.4±0.5‰ (n=4) and -39.8±0.2‰ (n=3), respectively.
238
Laboratory soil δ15N-NO measurements
239
To test the DFC-TEA method using real soil samples, approximately 4 kg of soil was
240
collected from the upper 10 cm of an urban forest soil in Pittsburgh, PA. Before use, soil
241
samples were sieved by passing through a 2-mm sieve and air-dried for 14 days. To
242
trigger NO pulses, 35 g of air-dried soil samples was added to the Teflon jar, mixed
243
thoroughly, and wetted by deionized water to achieve 100% water holding capacity. With
244
the continuous purging of the jar headspace, the soil samples were subject to drying-out
245
over the next 48 h, and NO was collected periodically for the δ15N-NO analysis.
ACS Paragon Plus Environment
Environmental Science & Technology
246
Field soil δ15N-NO measurements
247
To verify the DFC-TEA method under varying field conditions, the field DFC system
248
was deployed using the University of Pittsburgh Mobile Air Quality Laboratory to
249
measure δ15N-NO in a field soil rewetting experiment (see Figure S10 in the SI for the
250
field setup). A waterproof tarp (300 cm×240 cm) was erected over a fallow, urban plot in
251
Pittsburgh, PA for 2 weeks (8/15/2016 - 8/29/2016) to exclude precipitation inputs. After
252
the drying period, four soil plots were respectively wetted on four consecutive days using
253
500 mL of MilliQ water, 20 mM KNO3 (δ15N=46.5±0.3‰), 10 mM NaNO2
254
(δ15N=1.0±0.4‰), and 20 mM NH4Cl (δ15N=1.7±0.1‰) solutions. These N amendment
255
solutions were chosen because they are common precursors of NO in major NO-
256
producing processes.31, 36 Previous studies have reported that common δ15N values of N
257
fertilizers are not very different from 0‰ (e.g., -4.4‰ to 0.3‰),78 while δ15N values of
258
atmospherically deposited NO2, NO3- and NH4+ could vary over a wider range (e.g., -
259
10‰ to 15‰), depending on source contributions.42, 79 Hence, the δ15N values of the
260
amended NO2- and NH4+ are within the environmentally relevant range, whereas the δ15N
261
of the added NO3- is significant higher. The NO, NO2, and NH3 fluxes were continuously
262
measured before and after the soil rewetting, and NO was collected periodically for the
263
δ15N-NO analysis.
264 265
RESULTS AND DISCUSSION
266
We evaluate each step in the NO collection and report isotopic results that document the
267
overall precision and accuracy of the δ15N-NO analysis. We then present δ15N-NO
ACS Paragon Plus Environment
Page 12 of 44
Page 13 of 44
Environmental Science & Technology
268
measurements from laboratory and field soil rewetting experiments that demonstrate
269
utility of the DFC-TEA method for resolving soil NO dynamics.
270
NO conversion in excess O3
271
The reaction time of the inner tubing of the chemiluminescent analyzer and the reaction
272
tube plus the inner tubing were estimated to be 1.4 s and 6.4 s, respectively, resulting in a
273
reaction time of the reaction tube of 5 s at the measured flow temperature (22 °C) (Figure
274
S7 in the SI). This estimated reaction time is consistent with the residence time calculated
275
from the assumption of plug flow in the reaction tube. Based on this reaction time and the
276
average O3 concentration of 2911 ppbv, numerical model calculations including reactions
277
R1-R580 and NO3 loss on the interior tubing wall (R6 in Table S2 in the SI)81 indicate that
278
NO is quantitatively converted in the reaction tube and that the specific conversion of NO
279
to NO2 is between 98.7% and 99.0% over a wide range of NO concentrations (0-1000
280
ppbv) at 22 °C (Figure S8a in the SI). Notably, the remaining NO from the conversion
281
exists primarily as N2O5 (Figure S8b in the SI).
282
Deviations from controlled laboratory condition in the field may result in
283
variations in the modeled NO conversion efficiency.61 We therefore modeled the effects
284
of temperature variation and soil emission of biogenic volatile organic carbon (BVOC)82
285
on the NO conversion (reactions R7 and R8 in Table S2 in the SI)83. The results indicate
286
that the conversion of NO to NO2 is not likely to fall below 98% over a temperature
287
range of 0-40°C in conjunction with high BVOC emissions (e.g., 100 ppbv isoprene in
288
the chamber) (details on the extended modeling are given in the SI). In addition, slight
289
variations in the reaction time may result from changes in temperature and pressure of the
290
sample flow (e.g., pressure increase induced by the attachment of the gas washing bottle).
ACS Paragon Plus Environment
Environmental Science & Technology
291
While the effect of these variations on the NO conversion is difficult to empirically
292
quantify, any uncertainty in converting NO under the tested conditions is reflected in the
293
over method precision and accuracy for the δ15N-NO measurement.
294
NO2 collection in TEA solution
295
The 20% TEA solution was 100% efficient at collecting NO2. This was confirmed
296
by collecting a flow of reference NO2 at 1 ppmv using the laboratory DFC system (Table
297
1). Importantly, because the 20% TEA solution foams rigorously upon sparging, the
298
applied total flow rate (1.6 slpm of the sample flow plus 0.4 slpm of the O3 flow) was
299
chosen to avoid solution spill. We have also tested TEA solution from another brand
300
(BioUltra) that foams much less rigorously (coarse bubbles). However, consistent low
301
collection efficiency (0.05). The deviations from 100% NO recovery likely reflect inefficiencies in the NO
309
conversion (see above), the high uncertainty in the NO3- concentration determination
310
(e.g., for the 12 ppbv and 25 ppbv NO collection samples), and/or NO loss within the
311
system (e.g., NO loss in the HONO scrubbing solution and on the interior wall of the
312
field chamber). Importantly, the high and consistent NO recovery is a direct evidence that
ACS Paragon Plus Environment
Page 14 of 44
Page 15 of 44
Environmental Science & Technology
313
the sub-sampling was effective to minimize the TEA aging problem, if any, for a storage
314
time of at least 4 months.
315
For all the tank collection samples (n=52), about 90% of the collected NO or NO2
316
was in the form of NO2-, and the remainder as NO3- (Table 1). A 90% NO2-+10% NO3-
317
stoichiometry has been previously reported for active NO2 sampling using TEA-coated
318
cartridges.65 While a satisfactory explanation for the NO3- production cannot be given at
319
this time,65 the observed stoichiometry is best approximated by the redox reaction
320
between NO2 and TEA in the presence of water that gives a theoretical 1:1 conversion of
321
NO2 to NO2-.64, 65, 84 Although it is well known that N2O5 hydrolyzes in water as HNO3,85
322
which is preserved as NO3- in an alkaline TEA solution, whether N2O5 produced in the
323
NO conversion can be collected in the TEA solution as NO3- is not possible to quantify in
324
this case due to the high uncertainty in the NO3- concentration determination (i.e., ±0.36
325
µM) but will be the subject of future research.
326
It is worth noting that collection efficiency of the 20% TEA solution may be
327
subject to decrease over longer collection periods due to presence of O2, O3, and CO2 in
328
the sample flow that can compete with NO2 for TEA oxidation and decrease solution pH.
329
Given that the DFC-TEA method described here is developed to characterize transient
330
variations of soil NO emissions, use of 20% TEA solution for prolonged collection (i.e.,
331
>2 h) should be further investigated to ensure high and consistent collection efficiency.
332
Analytical uncertainty of the denitrifier method and the total N blank
333
The pooled standard deviation for each of the isotopic standards made in 20% TEA
334
solution and measured along with individual sample sets was: 0.3‰, 0.3‰, and 0.8‰ for
335
δ15N of IAEA-N3, USGS34, and RSIL20, respectively; 0.7‰ and 0.7‰ for δ18O of
ACS Paragon Plus Environment
Environmental Science & Technology
336
IAEA-N3 and USGS34, respectively; and 1.2‰ for ∆17O of USGS35. The lower
337
precision of the δ15N analysis of RSIL20 (0.8‰ relative to 0.3‰ for other standards) is
338
due to the larger uncertainty in measuring diluted RSIL20 solutions that require large
339
injection volumes (Figure 3a). To further understand this volume dependence, we
340
estimated the total N blank associated with the δ15N analysis of the TEA samples (i.e.,
341
TEA N blank + blank N associated with the denitrifier medium68, 71) by quantifying
342
shrinkage of the N isotope-ratio scale between USGS34 and RSIL20 measured in each
343
run of the TEA collection samples86 (more details are described in the SI). The results
344
show that the fractional blank size (fB) ranged between 0.04 and 0.18 across different runs
345
and was significantly, positively correlated with the sample volume and the measured
346
δ15N of RSIL20 (δ15NRSIL20-m) (Figure 3a). Fitting a linear equation to the molar amount
347
of the total N blank and the sample volume indicates that the N blank likely consisted of
348
a constant component of 0.46±0.12 nmol and a sample volume-dependent component of
349
0.23±0.06 nmol·mL-1 (Figure S5 in the SI; Figure 2); this is consistent with the blank size
350
estimated by injecting blank 20% TEA solution (details are provided in the SI). From the
351
linear relationship between fB and δ15NRSIL20-m, the δ15N of the blank N appears to be
352
~10‰ across different runs (Figure 3a). These consistent and predictable behaviors of the
353
total N blank indicate with a high degree of confidence that its isotope effect is implicitly
354
corrected during the δ15N analysis using the blank-matching strategy.
355
Isobaric interference from mass-independent oxygen isotopic composition
356
The δ18O of RSIL20 calibrated against IAEA-N3 and USGS34 ranged from -
357
25.8±0.9‰ to -22.7±1.5‰ across different runs, with an average of -23.7±1.1‰. This
358
results in an isotopic offset of about 28‰ between the measured apparent δ18O and the
ACS Paragon Plus Environment
Page 16 of 44
Page 17 of 44
Environmental Science & Technology
359
“true” δ18O (4.5‰) of RSIL20, in line with the branching fractionation between NO3- and
360
NO2- during denitrification (25-30‰).70 This implies that the oxygen isotopic exchange
361
between NO2- and water is limited in the alkaline TEA solution. Not surprisingly,
362
positive ∆17O values were observed in the N2O generated from collected samples. To
363
understand the transfer of the ∆17O anomaly from O3 during the NO conversion, a
364
theoretical ∆17O of the NO2 produced from the NO+O3 reaction (R1 in Table S2 in the
365
SI) was calculated to be 22.5±1.8‰ (details are provided in the SI). This theoretical ∆17O
366
value is not very different from the measured ∆17O values, which had an average of
367
19.7±0.8‰ across different runs (Table 1), indicating that the NO+O3 reaction essentially
368
dominated during the NO conversion. The measured ∆17O values led to a 1.0-1.2‰
369
correction of the measured δ15N values. For samples without sufficient concentrations for
370
∆17O measurement, the average ∆17O value (19.7‰) was used for the correction. This is
371
not a complete correction, in that the expression of the isobaric interference depends on fB
372
relative to each sample. Nevertheless, the resultant overcorrection on the δ15N of the low
373
concentration samples is 0.05; Figure S6b in the SI). The DFC-TEA method and the EPA NOx
383
collection method generally agree within 0.3‰, although discrepancies within individual
384
sets of collected samples ranged from -1.3‰ to 1.6‰. The largest discrepancies between
385
the two methods occurred with the lowest sample concentrations (i.e., 12 ppbv NO
386
collection samples) (Figure 3b). For these low concentration samples, isotopic analyses
387
were conducted on 5 nmol of NO2-+NO3- (achieved with a 3.8 mL injection of the
388
collection samples) and of a fB of 0.18. Therefore, although Pdiff did not correlate with the
389
sample concentration (Figure S6a in the SI), the collection samples with lower
390
concentrations were more prone to random error in matching blank between the standards
391
and samples due to their higher fB. Consequently, for accurate δ15N-NO analyses, soil NO
392
should be collected to achieve >3 µM NO2-+NO3- in the solution within 2 h (equivalent to
393
collecting a flow of >26 ppbv NO over a 2 hour period). Blank 20% TEA solution should
394
then be used to dilute both soil NO collection samples and isotopic standards within a
395
batch to a common concentration for injection of 10 nmol of N using the denitrifier
396
method. Control tests using a soil NO sample collected from the laboratory rewetting
397
experiment (δ15N = -37.1‰, [NO2-]+[NO3-] = 9.2 µM) indicate that dilution-induced
398
uncertainty in the δ15N was 3 µM
399
NO2-+NO3- in the solution (data not shown). This uncertainty is within the analytical
400
uncertainty (i.e., ±0.8‰ for RSIL20). Therefore, we always group samples with similar
401
concentrations such that the dilution factor does not exceed 3.
402
Overall, our inter-calibration effort demonstrates that although the NO recovery
403
was slightly less than 100%, fractionation during chamber mixing and NO conversion
404
and collection is effectively minimized under the tested conditions. The derived standard
ACS Paragon Plus Environment
Page 18 of 44
Page 19 of 44
Environmental Science & Technology
405
deviation of ±1.1‰ based on all the collection samples with an average Pdiff of ±7% (i.e.,
406
sample peak area is within 100±7% of that of RSIL20) represents the overall accuracy
407
and precision across the entire method, accounting for propagated errors from the total N
408
blank and its mismatch between the standards and samples. While the method precision is
409
lower than that of the modified EPA method (Table S1 in the SI), our integrated method
410
featuring simultaneous NO flux measurement and collection is the first to show its
411
suitability for unbiased soil δ15N-NO determination under realistic, varying soil
412
conditions. Furthermore, the method is more convenient than previous methods and does
413
not require time-consuming pretreatments for δ15N analysis (Table S1 in the SI). Given
414
the good result from the inter-calibration, the tank NO can be utilized as a secondary
415
standard for correcting the isobaric interference. For instance, the tank NO can be
416
collected before and after soil NO collection; ∆17O of the tank collection samples can
417
then be estimated using an empirical relationship scaling a 1‰ increase from accepted
418
δ15N value (-71.4‰ in this case) to every 18.8‰ increase in ∆17O to correct the soil
419
collection samples.86
420
Application to pulsed soil NO emissions
421
Pulsed NO emission was triggered by soil rewetting under both the laboratory and field
422
conditions (Figure 4). In the laboratory, the pulsed NO emission had evident temporal
423
variations with a rapid initial NO pulse being triggered upon the rewetting (Figure 4a).
424
While the initial NO pulse was absent under the field conditions, possibly due to the
425
relatively high pre-wetting soil water content (0.17 cm3·cm-3), the rewetting and N
426
amendments caused significantly increased NO emission as compared to the pre-wetting
ACS Paragon Plus Environment
Environmental Science & Technology
427
emission (47±16 nmol·m-2·min-1). Particularly, a dramatic increase in NO emission was
428
triggered by the NO2- addition (Figure 4b).
429
Twenty and 15 samples were collected for the δ15N-NO analysis from the
430
laboratory and field experiments, respectively (Table S5 and S6 in the SI). The average
431
NO recovery was 102.2±5.6% and 108.6±11.0% for the laboratory and field collection
432
samples, respectively. The >100% recovery was detected mostly in samples collected
433
under the NO2- addition (NO recovery = 117.5±11.6%; Table S6 in the SI). We suspect
434
that the >100% NO recovery might result from our underestimation of the soil NO
435
emission due to the slow response time of the chemiluminescent analyzer (>30 s),
436
especially given that transient fluctuations in the NO flux were likely triggered by the
437
NO2- addition (Figure 4b). Alternatively, the >100% recovery could result from soil
438
emission of NOy (NOy = NO2 + HONO + HNO3 + other non-NO reactive N oxides),87
439
that can potentially be collected in the TEA solution as NO2- and/or NO3-.65 If soil NOy
440
emission was significant during measurement, it would be detected as NO2 by our
441
chemiluminescent analyzer with a molybdenum convertor.88 Because the NO2 flux never
442
exceed 2% of the simultaneous NO flux (Figure S11 and Figure S12 in the SI),
443
contributions of NOy emission to the NO recovery and the measured δ15N-NO are
444
considered negligible. Future application of the DFC-TEA method can be coupled to a
445
faster NO measurement system and existing denuder and wet chemistry methods76, 89 that
446
quantitatively scrub NOy without significant loss of NO.
447
The measured soil δ15N-NO exhibited intriguing patterns that are indicative of
448
mechanisms underlying the soil NO emissions (Figure 4). In the laboratory rewetting of
449
the air-dried soil samples, the initial NO pulse had higher δ15N values (-36.7~-39.9‰)
ACS Paragon Plus Environment
Page 20 of 44
Page 21 of 44
Environmental Science & Technology
450
than NO emission after 12 h post-wetting (-52.0~-53.6‰; Figure 4a). Recent work by
451
Homyak et al.33 provided evidence that in arid soils, abiotic reactions govern the rapid
452
initial NO pulse, whereas microbial processes control later emissions as microbes recover
453
from drought stress. Therefore, the higher δ15N-NO values associated with the initial NO
454
pulse may suggest that abiotic reactions likely bear a smaller isotopic fractionation on
455
NO production than microbial processes. However, the temporal variation of δ15N-NO
456
could also result from changing rates of microbial NO production90 or sequential
457
resuscitation of different microbial groups91 during the rewetting. Further constraints on
458
the relevant isotope effects are needed to tease apart the relative importance of abiotic
459
and microbial pathways sustaining pulsed NO emissions in soils. In the field rewetting
460
experiment where an initial NO pulse was lacking, the measured soil δ15N-NO responded
461
differently to the added N precursors (Figure 4b). First, the δ15N-NO values that evolved
462
from the NH4+ (-59.8‰ to -56.0‰) and NO2- (-34.4‰ to -23.4‰) amendments were
463
significantly lower and higher relative to the control (MilliQ water addition; -44.3‰ to -
464
41.3‰), respectively, in spite of the almost equal δ15N of the added NH4+ and NO2-.
465
Secondly, despite the high δ15N of the added NO3- (i.e., 46.5‰), the δ15N-NO values
466
measured from the NO3- amendment (-40.7‰ to -39.4‰) were not significantly different
467
from those in the control. The measured soil δ15N-NO and its differential responses to the
468
amended N sources indicate that various soil NO-producing processes (e.g., nitrification,
469
denitrification, and chemodenitrification), stimulated by different N amendments, likely
470
bear distinguishable isotopic imprints on NO production, similar to what has been
471
observed in soil nitrous oxide (N2O) studies.37-39 For example, N2O production in soil
472
was found to be associated with a larger isotope effect for nitrification of NH4+ (e.g., -
ACS Paragon Plus Environment
Environmental Science & Technology
473
45‰ to -67‰) than denitrification of NO2- (e.g., -35‰ to -22‰).37-39 Thus, soil δ15N-NO
474
measurement could potentially provide important implications for understanding
475
couplings between soil NO and N2O emissions, in that NO is the precursor of N2O in
476
most abiotic and microbial processes.10, 17-20 Finally, the measured soil δ15N-NO values
477
are significantly lower than other measured NOx emissions sources,44, 45, 47-49 confirming
478
use of soil δ15N-NO as a robust tracer of regional N deposition.41, 42 Quantification of
479
isotope effects associated with NO dynamics in soils therefore represents an important
480
avenue for future research on the soil-atmosphere cycling of reactive N.
481 482
ACKNOWLEDGEMENT
483
This material is based upon work supported by a National Science Foundation CAREER
484
award (Grant No. 1253000) to EME. We thank Kathrine Redling, Vivian Feng, Madeline
485
Ellgass, and Madeline Gray for assistance with isotopic measurements.
486 487
SUPPORTING INFORMATION AVAILABLE
488
Supporting information includes a description of: (1) comparison between the DFC-TEA
489
method and previously published NOx collection methods; (2) the dynamic flux chamber
490
system; (3) the protocol for NO2- and NO3- concentration measurement using the
491
modified spongy cadmium reduction method; (4) the total N blank and the blank-
492
matching strategy; (5) extended modeling of the NO conversion in excess O3; (6)
493
calculation of the theoretical ∆17O value of the NO2 produced from the NO+O3 reaction;
494
(7) experimental setup of the field rewetting experiment; (8) soil NOy emissions in the
ACS Paragon Plus Environment
Page 22 of 44
Page 23 of 44
Environmental Science & Technology
495
laboratory and field soil rewetting experiments; and (9) complete datasets of the
496
laboratory and field collection samples.
497 498
REFERENCES CITED
499
(1) Climate change 2013: the physical science basis. Contribution of working group I to
500
the fifth assessment report of the Intergovernmental Panel on Climate Change. 2013.
501 502
(2) Richter, A.; Burrows, J. P.; Nüß, H.; Granier, C.; Niemeier, U. Increase in
503
tropospheric nitrogen dioxide over China observed from space. Nature 2007, 437(7055),
504
129-132.
505 506
(3) Jacob, D. J.; Heikes, E. G.; Fan, S. M.; Logan, J. A.; Mauzerall, D. L.; Bradshaw, J.
507
D.; Singh, H. B.; Gregory, G. L.; Talbot, R. W.; Blake, D. R.; Sachse, G. W. Origin of
508
ozone and NOx in the tropical troposphere: A photochemical analysis of aircraft
509
observations over the South Atlantic basin. J. Geophys. Res. Atmos. 1996, 101(D19),
510
24235-24250.
511 512
(4) Jang, M.; Czoschke, N. M.; Lee, S.; Kamens, R. M. Heterogeneous atmospheric
513
aerosol production by acid-catalyzed particle-phase reactions. Science 2002, 298(5594),
514
814-817.
515 516
(5) Likens, G. E.; Driscoll, C. T.; Buso, D. C. Long-term effects of acid rain: response
517
and recovery of a forest ecosystem. Science 1996, 272(5259), 244.
518
ACS Paragon Plus Environment
Environmental Science & Technology
519
(6) Akimoto, H. Global air quality and pollution. Science, 2003, 302(5651), 1716-1719.
520 521
(7) Jaeglé, L.; Steinberger, L.; Martin, R. V.; Chance, K. Global partitioning of NOx
522
sources using satellite observations: Relative roles of fossil fuel combustion, biomass
523
burning and soil emissions. Faraday discuss. 2005, 130, 407-423.
524 525
(8) Zhang, R.; Tie, X.; Bond, D.W. Impacts of anthropogenic and natural NOx sources
526
over the US on tropospheric chemistry. PNAS 2003, 100(4), 1505-1509.
527 528
(9) Galbally, I. E.; Roy, C. R. Loss of fixed nitrogen from soils by nitric oxide exhalation.
529
Nature 1978, 275, 734-735.
530 531
(10) Skiba, U.; Smith, K. A. Nitrification and denitrification as sources of nitric oxide
532
and nitrous oxide in a sandy loam soil. Soil Biol. Biochem. 1993, 25(11), 1527-1536.
533 534
(11) Yienger, J. J.; Levy, H. Empirical model of global soil‐biogenic NOx emissions. J.
535
Geophys. Res. Atoms. 1995, 100(D6), 11447-11464.
536 537
(12) Bertram, T. H.; Heckel, A.; Richter, A.; Burrows, J. P.; Cohen, R. C. Satellite
538
measurements of daily variations in soil NOx emissions. Geophys. Res. Lett. 2005, 32(24),
539
L24812.
540
ACS Paragon Plus Environment
Page 24 of 44
Page 25 of 44
Environmental Science & Technology
541
(13) Hudman, R. C.; Russell, A. R.; Valin, L. C.; Cohen, R. C. Interannual variability in
542
soil nitric oxide emissions over the United States as viewed from space. Atmos. Chem.
543
Phys. 2010, 10(20), 9943-9952.
544 545
(14) Steinkamp, J.; Lawrence, M. G. Improvement and evaluation of simulated global
546
biogenic soil NO emissions in an AC-GCM. Atmos. Chem. Phys. 2011, 11(12), pp.6063-
547
6082.
548 549
(15) Vinken, G. C. M.; Boersma, K. F.; Maasakkers, J. D.; Adon, M.; Martin, R. V.
550
Worldwide biogenic soil NOx emissions inferred from OMI NO2 observations. Atmos.
551
Chem. Phys. 2014, 14(18), 10363-10381.
552 553
(16) Steinkamp, J.; Ganzeveld, L. N.; Wilcke, W.; Lawrence, M. G. Influence of
554
modelled soil biogenic NO emissions on related trace gases and the atmospheric
555
oxidizing efficiency. Atmos. Chem. Phys. 2009, 9(8), 2663-2677.
556 557
(17) Zumft, W. G. Cell biology and molecular basis of denitrification. Microbiol. Mol.
558
Bio. Rev. 1997, 61(4), 533-616.
559 560
(18) Kester, R. A.; De Boer, W.; Laanbroek, H. J. Production of NO and N2O by Pure
561
Cultures of Nitrifying and Denitrifying Bacteria during Changes in Aeration. Appl.
562
Environ. Microbiol. 1997, 63(10), 3872-3877.
563
ACS Paragon Plus Environment
Environmental Science & Technology
564
(19) Firestone, M. K.; Davidson, E. A. Microbiological basis of NO and N2O production
565
and consumption in soil. In Exchange of trace gases between terrestrial ecosystems and
566
the atmosphere; Andreae, M. O.; Schimel, D. S.; Robertson, G. P. Eds.; John Wiley &
567
Sons Ltd: Berlin 1989; pp 7-21.
568 569
(20) Venterea, R. T.; Rolston, D. E. Nitric and nitrous oxide emissions following
570
fertilizer application to agricultural soil: Biotic and abiotic mechanisms and kinetics. J.
571
Geophys. Res. Atoms. 2000, 105(D12), 15117-15129.
572 573
(21) McCalley, C. K.; Sparks, J. P. Abiotic gas formation drives nitrogen loss from a
574
desert ecosystem. Science 2009, 326(5954), 837-840.
575 576
(22) Homyak, P. M.; Kamiyama, M.; Sickman, J. O.; Schimel, J. P. Acidity and organic
577
matter promote abiotic nitric oxide production in drying soils. Global Change Biol. 2016,
578
23(4), 1735-1747.
579 580
(23) Yang, W. X.; Meixner, F. X. Laboratory studies on the release of nitric oxide from
581
sub-tropical grassland soils: The effect of soil temperature and moisture. In Gaseous
582
Nitrogen emissions from grasslands; Jarvis, S. C.; Pain, B. F. Eds.; Cab International:
583
1997; pp 67-71.
584
ACS Paragon Plus Environment
Page 26 of 44
Page 27 of 44
Environmental Science & Technology
585
(24) Van Dijk, S. M.; Gut, A.; Kirkman, G. A.; Gomes, B. M.; Meixner, F. X.; Andreae,
586
M. O. Biogenic NO emissions from forest and pasture soils: Relating laboratory studies
587
to field measurements. J. Geophys. Res. Atmos. 2002, 107(D20).
588 589
(25) Hall, S. J.; Matson, P. A. Nitrogen oxide emissions after nitrogen additions in
590
tropical forests. Nature 1999, 400(6740), 152-155.
591 592
(26) Davidson, E. A.; Keller, M.; Erickson, H. E.; Verchot, L. V.; Veldkamp, E. Testing a
593
Conceptual Model of Soil Emissions of Nitrous and Nitric Oxides. Bioscience 2000,
594
50(8), 667-680.
595 596
(27) Davidson, E. A.; Verchot, L. V. Testing the Hole‐in‐the‐Pipe Model of nitric and
597
nitrous oxide emissions from soils using the TRAGNET database. Global Biogeochem.
598
Cycles 2000, 14(4), 1035-1043.
599 600
(28) Hudman, R. C.; Moore, N. E.; Mebust, A. K.; Martin, R. V.; Russell, A. R.; Valin, L.
601
C.; Cohen, R. C. Steps towards a mechanistic model of global soil nitric oxide emissions:
602
implementation and space based-constraints. Atmos. Chem. Phys. 2012, 12(16), 7779-
603
7795.
604 605
(29) Jaeglé, L.; Martin, R. V.; Chance, K.; Steinberger, L.; Kurosu, T. P.; Jacob, D. J.;
606
Modi, A. I.; Yoboué, V.; Sigha‐Nkamdjou, L.; Galy‐Lacaux, C. Satellite mapping of
ACS Paragon Plus Environment
Environmental Science & Technology
607
rain‐induced nitric oxide emissions from soils. J. Geophys. Res. Atmos. 2004, 109(D21),
608
D21310.
609 610
(30) Oikawa, P. Y.; Ge, C.; Wang, J.; Eberwein, J. R.; Liang, L. L.; Allsman, L. A.;
611
Grantz, D. A.; Jenerette, G. D. Unusually high soil nitrogen oxide emissions influence air
612
quality in a high-temperature agricultural region. Nat. Commun. 2015, 6, 8753.
613 614
(31) Davidson, E. A. Sources of nitric oxide and nitrous oxide following wetting of dry
615
soil. Soil Sci. Soc. Am. J. 1992, 56(1), 95-102.
616 617
(32) Davidson, E. A. Pulses of nitric oxide and nitrous oxide flux following wetting of
618
dry soil: An assessment of probable sources and importance relative to annual
619
fluxes. Ecol. Bull. 1992, 149-155.
620 621
(33) Homyak, P. M.; Blankinship, J. C.; Marchus, K.; Lucero, D. M.; Sickman, J. O.;
622
Schimel, J. P. Aridity and plant uptake interact to make dryland soils hotspots for nitric
623
oxide (NO) emissions. PNAS 2016, 113(19), E2608-E2616.
624 625
(34) Wang, Y.; McElroy, M. B.; Martin, R. V.; Streets, D. G.; Zhang, Q.; Fu, T. M.
626
Seasonal variability of NOx emissions over east China constrained by satellite
627
observations: Implications for combustion and microbial sources. J. Geophys. Res.
628
Atmos. 2007, 112(D6).
629
ACS Paragon Plus Environment
Page 28 of 44
Page 29 of 44
Environmental Science & Technology
630
(35) Hudman, R. C.; Jacob, D. J.; Turquety, S.; Leibensperger, E. M.; Murray, L. T.; Wu,
631
S.; Gilliland, A. B.; Avery, M.; Bertram, T. H.; Brune, W.; Cohen, R. C. Surface and
632
lightning sources of nitrogen oxides over the United States: Magnitudes, chemical
633
evolution, and outflow. J. Geophys. Res. Atmos. 2007, 112(D12).
634 635
(36) Denk, T. R.; Mohn, J.; Decock, C.; Lewicka-Szczebak, D.; Harris, E.; Butterbach-
636
Bahl, K.; Kiese, R.; Wolf, B. The nitrogen cycle: A review of isotope effects and isotope
637
modeling approaches. Soil Biol. Biochem. 2017, 105, 121-137.
638 639
(37) Mariotti, A.; Germon, J. C.; Hubert, P.; Kaiser, P.; Letolle, R.; Tardieux, A.;
640
Tardieux, P. Experimental determination of nitrogen kinetic isotope fractionation: some
641
principles; illustration for the denitrification and nitrification processes. Plant Soil 1981,
642
62(3), 413-430.
643 644
(38) Sutka, R. L.; Ostrom, N. E.; Ostrom, P. H.; Breznak, J. A.; Gandhi, H.; Pitt, A. J.; Li,
645
F. Distinguishing nitrous oxide production from nitrification and denitrification on the
646
basis of isotopomer abundances. Appl. Environ. Microbio. 2006, 72(1), 638-644.
647 648
(39) Park, S.; Pérez, T.; Boering, K. A.; Trumbore, S. E.; Gil, J.; Marquina, S.; Tyler, S.
649
C. Can N2O stable isotopes and isotopomers be useful tools to characterize sources and
650
microbial pathways of N2O production and consumption in tropical soils?. Global
651
Biogeochem. Cycles 2011, 25(1), GB1001.
652
ACS Paragon Plus Environment
Environmental Science & Technology
653
(40) Hastings, M. G.; Casciotti, K. L.; Elliott, E. M. Stable isotopes as tracers of
654
anthropogenic nitrogen sources, deposition, and impacts. Elements 2013, 9(5), 339-344.
655 656
(41) Elliott, E. M.; Kendall, C.; Wankel, S. D.; Burns, D. A.; Boyer, E. W.; Harlin, K.;
657
Bain, D. J.; Butler, T. J. Nitrogen isotopes as indicators of NOx source contributions to
658
atmospheric nitrate deposition across the midwestern and northeastern United
659
States. Environ. Sci. Technol. 2007, 41(22), 7661-7667.
660 661
(42) Elliott, E. M.; Kendall, C.; Boyer, E. W.; Burns, D. A.; Lear, G. G.; Golden, H. E.;
662
Harlin, K.; Bytnerowicz, A.; Butler, T. J.; Glatz, R. Dual nitrate isotopes in dry
663
deposition: Utility for partitioning NOx source contributions to landscape nitrogen
664
deposition. J. Geophys. Res. Biogeosci. 2009, 114(G4), G04020.
665 666
(43) Hastings, M. G.; Jarvis, J. C.; Steig, E. J. Anthropogenic impacts on nitrogen
667
isotopes of ice-core nitrate. Science 2009, 324(5932), 1288-1288.
668 669
(44) Felix, J. D.; Elliott, E. M.; Shaw, S. L. Nitrogen isotopic composition of coal-fired
670
power plant NOx: influence of emission controls and implications for global emission
671
inventories. Environ. Sci. Technol. 2012, 46(6), 3528-3535.
672 673
(45) Redling, K.; Elliott, E.; Bain, D.; Sherwell, J. Highway contributions to reactive
674
nitrogen deposition: tracing the fate of vehicular NOx using stable isotopes and plant
675
biomonitors. Biogeochemistry 2013, 116(1-3), 261-274.
ACS Paragon Plus Environment
Page 30 of 44
Page 31 of 44
Environmental Science & Technology
676 677
(46) Hoering, T. The isotopic composition of the ammonia and the nitrate ion in
678
rain. Geochim. Cosmochim. Acta 1957, 12(1), 97-102.
679 680
(47) Walters, W. W.; Goodwin, S. R.; Michalski, G. Nitrogen stable isotope composition
681
(δ15N) of vehicle-emitted NOx. Environ. Sci. Technol. 2015, 49(4), 2278-2285.
682 683
(48) Walters, W. W.; Tharp, B. D.; Fang, H.; Kozak, B. J.; Michalski, G. Nitrogen
684
isotope composition of thermally produced NOx from various fossil-fuel combustion
685
sources. Environ. Sci. Technol. 2015, 49(19), 11363-11371.
686 687
(49) Fibiger, D. L.; Hastings, M. G. First measurements of the nitrogen isotopic
688
composition of NOx from biomass burning. Environ. Sci. Technol. 2016, 50(21), 11569-
689
11574.
690 691
(50) Li, D.; Wang, X. Nitrogen isotopic signature of soil-released nitric oxide (NO) after
692
fertilizer application. Atmos. Environ. 2008, 42(19), 4747-4754.
693 694
(51) Felix, J. D.; Elliott, E. M. The agricultural history of human‐nitrogen interactions as
695
recorded in ice core δ15N‐NO3-. Geophys. Res. Lett. 2013, 40(8), 1642-1646.
696
ACS Paragon Plus Environment
Environmental Science & Technology
697
(52) Felix, J. D.; Elliott, E.M. Isotopic composition of passively collected nitrogen
698
dioxide emissions: Vehicle, soil and livestock source signatures. Atmos. Environ. 2014,
699
92, 359-366.
700 701
(53) Hutchinson, G. L.; Yang, W. X.; Andre, C. E. Overcoming humidity dependence of
702
the chromium trioxide converter used in luminol-based nitric oxide detection. Atmos.
703
Environ. 1998, 33(1), 141-145.
704 705
(54) Robinson, J. K.; Bollinger, M. J.; Birks, J. W. Luminol/H2O2 chemiluminescence
706
detector for the analysis of nitric oxide in exhaled breath. Anal. Chem. 1999, 71(22),
707
5131-5136.
708 709
(55) Fibiger, D. L.; Hastings, M. G.; Lew, A. F.; Peltier, R. E. Collection of NO and NO2
710
for isotopic analysis of NOx emissions. Anal. Chem. 2014, 86(24), 12115-12121.
711 712
(56) Wojtal, P. K.; Miller, D. J.; O'Conner, M.; Clark, S. C.; Hastings, M. G. Automated,
713
high-resolution mobile collection system for the nitrogen isotopic analysis of NOx. J. Vis.
714
Exp. 2016, 118, e54962.
715 716
(57) Pape, L.; Ammann, C.; Nyfeler-Brunner, A.; Spirig, C.; Hens, K.; Meixner, F. X. An
717
automated dynamic chamber system for surface exchange measurement of non-reactive
718
and reactive trace gases of grassland ecosystems. Biogeosciences, 2009, 6(3), 405-429.
719
ACS Paragon Plus Environment
Page 32 of 44
Page 33 of 44
Environmental Science & Technology
720
(58) Yu, Z.; Slater, L. D.; Schäfer, K. V.; Reeve, A. S.; Varner, R. K. Dynamics of
721
methane ebullition from a peat monolith revealed from a dynamic flux chamber system. J.
722
Geophys. Res. Biogeosci. 2014, 119(9), 1789-1806.
723 724
(59) Miyazaki, K.; Matsumoto, J.; Kato, S.; Kajii, Y. Development of atmospheric NO
725
analyzer by using a laser-induced fluorescence NO2 detector. Atmos. Environ. 2008,
726
42(33), 7812-7820.
727 728
(60) Hargrove, J.; Zhang, J. Measurements of NOx, acyl peroxynitrates, and NOy with
729
automatic interference corrections using a NO2 analyzer and gas phase titration. Rev. Sci.
730
Instrum. 2008, 79(4), pp.046109-046109.
731 732
(61) Fuchs, H.; Dubé, W. P.; Lerner, B. M.; Wagner, N. L.; Williams, E. J.; Brown, S. S.
733
A sensitive and versatile detector for atmospheric NO2 and NOx based on blue diode laser
734
cavity ring-down spectroscopy. Environ. Sci. Technol. 2009, 43(20), 7831-7836.
735 736
(62) Wild, R. J.; Edwards, P. M.; Dubé, W. P.; Baumann, K.; Edgerton, E. S.; Quinn, P.
737
K.; Roberts, J. M.; Rollins, A. W.; Veres, P. R.; Warneke, C.; Williams, E. J. A
738
Measurement of Total Reactive Nitrogen, NOy, together with NO2, NO, and O3 via
739
Cavity Ring-down Spectroscopy. Environ. Sci. Technol. 2014, 48(16), 9609-9615.
740
ACS Paragon Plus Environment
Environmental Science & Technology
741
(63) Wagner, N. L.; Dubé, W. P.; Washenfelder, R. A.; Young, C. J.; Pollack, I. B.;
742
Ryerson, T. B.; Brown, S. S. Diode laser-based cavity ring-down instrument for NO3,
743
N2O5, NO, NO2 and O3 from aircraft. Atmos. Meas. Tech. 2011, 4(6), 1227-1240.
744 745
(64) Glasius, M.; Carlsen, M. F.; Hansen, T. S.; Lohse, C. Measurements of nitrogen
746
dioxide on Funen using diffusion tubes. Atmos. Environ. 1999, 33(8), 1177-1185.
747 748
(65) Cape, J. N. The use of passive diffusion tubes for measuring concentrations of
749
nitrogen dioxide in air. Crit. Rev. Anal. Chem. 2009, 39(4), 289-310.
750 751
(66) Wei, Y.; Oshima, M.; Simon, J.; Motomizu, S. The application of the
752
chromatomembrane cell for the absorptive sampling of nitrogen dioxide followed by
753
continuous determination of nitrite using a micro-flow injection system. Talanta, 2002,
754
57(2), 355-364.
755 756
(67) Jones, M. N. Nitrate reduction by shaking with cadmium: alternative to cadmium
757
columns. Water Res. 1984, 18(5), 643-646.
758 759
(68) Sigman, D. M.; Casciotti, K. L.; Andreani, M.; Barford, C.; Galanter, M. B. J. K.;
760
Böhlke, J. K. A bacterial method for the nitrogen isotopic analysis of nitrate in seawater
761
and freshwater. Anal. Chem. 2001, 73(17), 4145-4153.
762
ACS Paragon Plus Environment
Page 34 of 44
Page 35 of 44
Environmental Science & Technology
763
(69) Casciotti, K. L.; Sigman, D. M.; Hastings, M. G.; Böhlke, J. K.; Hilkert, A.
764
Measurement of the oxygen isotopic composition of nitrate in seawater and freshwater
765
using the denitrifier method. Anal. Chem. 2002, 74(19), 4905-4912.
766 767
(70) Casciotti, K. L.; Böhlke, J. K.; McIlvin, M. R.; Mroczkowski, S. J.; Hannon, J. E.
768
Oxygen isotopes in nitrite: analysis, calibration, and equilibration. Anal. Chem. 2007,
769
79(6), 2427-2436.
770 771
(71) McIlvin, M. R.; Casciotti, K. L. Technical updates to the bacterial method for nitrate
772
isotopic analyses. Anal. Chem. 2011, 83(5), 1850-1856.
773 774
(72) Michalski, G.; Savarino, J.; Böhlke, J. K.; Thiemens, M. Determination of the total
775
oxygen isotopic composition of nitrate and the calibration of a ∆17Ο nitrate reference
776
material. Anal. Chem. 2002, 74(19), 4989-4993.
777 778
(73) Kaiser, J.; Hastings, M. G.; Houlton, B. Z.; Röckmann, T.; Sigman, D. M. Triple
779
oxygen isotope analysis of nitrate using the denitrifier method and thermal decomposition
780
of N2O. Anal. Chem. 2007, 79(2), 599-607.
781 782
(74) Su, H.; Cheng, Y.; Oswald, R.; Behrendt, T.; Trebs, I.; Meixner, F. X.; Andreae, M.
783
O.; Cheng, P.; Zhang, Y.; Pöschl, U. Soil nitrite as a source of atmospheric HONO and
784
OH radicals. Science 2011, 333(6049), 1616-1618.
785
ACS Paragon Plus Environment
Environmental Science & Technology
786
(75) Oswald, R.; Behrendt, T.; Ermel, M.; Wu, D.; Su, H.; Cheng, Y.; Breuninger, C.;
787
Moravek, A.; Mougin, E.; Delon, C.; Loubet, B. HONO emissions from soil bacteria as a
788
major source of atmospheric reactive nitrogen. Science 2013, 341(6151), 1233-1235.
789 790
(76) Zhou, X.; Qiao, H.; Deng, G.; Civerolo, K. A method for the measurement of
791
atmospheric HONO based on DNPH derivatization and HPLC analysis. Environ. Sci.
792
Technol. 1999, 33(20), 3672-3679.
793 794
(77) Method 7 – Determination of nitrogen oxide emissions from stationary sources; US
795
Environmental Protection Agency; www3.epa.gov/ttn/emc/promgate/m-07.pdf.
796 797
(78) Michalski, G.; Kolanowski, M.; Riha, K.M. Oxygen and nitrogen isotopic
798
composition of nitrate in commercial fertilizers, nitric acid, and reagent salts. Isotopes
799
Environ. Health Stud. 2015, 51(3), 382-391.
800 801
(79) Altieri, K.E.; Hastings, M.G.; Peters, A.J.; Oleynik, S.; Sigman, D.M. Isotopic
802
evidence for a marine ammonium source in rainwater at Bermuda. Glob. Biogeochem.
803
Cycles 2014, 28(10), 1066-1080.
804 805
(80) Chemical Kinetics and Photochemical Data for Use in Atmospheric Studies
806
Evaluation Number 15; Jet Propulsion Laboratory, National Aeronautics and Space
807
Administration: Pasadena, CA, 2006; https://trs.jpl.nasa.gov/handle/2014/41648.
808
ACS Paragon Plus Environment
Page 36 of 44
Page 37 of 44
Environmental Science & Technology
809
(81) Dubé, W. P.; Brown, S. S.; Osthoff, H. D.; Nunley, M. R.; Ciciora, S. J.; Paris, M.
810
W.; McLaughlin, R. J.; Ravishankara, A. R. Aircraft instrument for simultaneous, in situ
811
measurement of NO3 and N2O5 via pulsed cavity ring-down spectroscopy. Rev. Sci.
812
Instrum. 2006, 77(3), 034101.
813 814
(82) Atkinson, R.; Arey, J. Gas-phase tropospheric chemistry of biogenic volatile organic
815
compounds: a review. Atmos. Environ. 2003, 37, 197-219.
816 817
(83) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Crowley, J. N.; Hampson, R. F.; Hynes, R.
818
G.; Jenkin, M. E.; Rossi, M. J.; Troe, J.; Subcommittee, I. U. P. A. C. Evaluated kinetic
819
and photochemical data for atmospheric chemistry: Volume II–gas phase reactions of
820
organic species. Atmos. Chem. Phys. 2006, 6(11), 3625-4055.
821 822
(84) Dahal, B.; Hastings, M. G. Technical considerations for the use of passive samplers
823
to quantify the isotopic composition of NOx and NO2 using the denitrifier method. Atmos.
824
Environ. 2016, 143, 60-66.
825 826
(85) Riemer, N.; Vogel, H.; Vogel, B.; Anttila, T.; Kiendler‐Scharr, A.; Mentel, T. F.
827
Relative importance of organic coatings for the heterogeneous hydrolysis of N2O5 during
828
summer in Europe. J. Geophys. Res. Atmos. 2009, 114(D17), D17307.
829
ACS Paragon Plus Environment
Environmental Science & Technology
830
(86) Coplen, T. B.; Böhlke, J. K.; Casciotti, K. L. Using dual‐bacterial denitrification to
831
improve δ15N determinations of nitrates containing mass‐independent 17O. Rapid
832
Commun. Mass Spectrom. 2004, 18(3), 245-250.
833 834
(87) Soper, F. M.; Boutton, T. W.; Groffman, P. M.; Sparks, J. P. Nitrogen trace gas
835
fluxes from a semiarid subtropical savanna under woody legume encroachment. Global
836
Biogeochem. Cycles 2016, 30(5), 614-628.
837 838
(88) Dunlea, E. J.; Herndon, S. C.; Nelson, D. D.; Volkamer, R. M.; San Martini, F.;
839
Sheehy, P. M.; Zahniser, M. S.; Shorter, J. H.; Wormhoudt, J. C.; Lamb, B. K.; Allwine,
840
E. J. Evaluation of nitrogen dioxide chemiluminescence monitors in a polluted urban
841
environment. Atmos. Chem. Phys. 2007, 7(10), 2691-2704.
842 843
(89) De Santis, F.; Allegrini, I.; Di Filippo, P.; Pasella, D. Simultaneous determination of
844
nitrogen dioxide and peroxyacetyl nitrate in ambient atmosphere by carbon-coated
845
annular diffusion denuder. Atmos. Environ. 1996, 30(14), 2637-2645.
846 847 848
(90) Mariotti, A.; Leclerc, A.; Germon, J. C. Nitrogen isotope fractionation associated
849
with the NO2-→N2O step of denitrification in soils. Can. J. Soil Sci. 1982, 62(2), 227-241.
850
ACS Paragon Plus Environment
Page 38 of 44
Page 39 of 44
Environmental Science & Technology
851
(91) Placella, S. A.; Brodie, E. L.; Firestone, M. K. Rainfall-induced carbon dioxide
852
pulses result from sequential resuscitation of phylogenetically clustered microbial
853
groups. PNAS 2012, 109(27), 10931-10936.
854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872
ACS Paragon Plus Environment
Environmental Science & Technology
873 874 875
Page 40 of 44
Table 1. Summary of the reference NO and NO2 tank collection using the DFC-TEA method under varying environmental conditions. The complete dataset is given in Table S4 in the SI.a RH (%)
NO2-+NO3(µM)
Recoveryb (%)
NO2- percent (%)
Pdiff (%)
δ15Nc (‰)
25.3
132.5 4.7
101.4 3.6
87.4 0.3
3.3 5.1
-40.1 0.8
44.6 27.1 34.2
1.4 4.1 11.9
95.0 100.7 98.2
97.0 93.1 94.0
-0.7 -1.2 0.7
-73.0 (-71.7) -70.3 (-69.2) -71.0 (-69.9)
18.8
749 ppbv NO (n=4) 120 22.8 47.5 14.2 NO collection – laboratory DFC system – temperature effect 34 ppbv NO (n=4) 120 11.5 92.0 4.0 101 ppbv NO (n=4) 120 30.8 28.8 11.8 NO collection – laboratory DFC system – interference 34 ppbv NO 120 23.0 33.1 4.0 + 500 ppbv NH3 (n=3) 101 ppbv NO 120 23.1 46.5 11.7 + 500 ppbv NH3 (n=4) 101 ppbv NO 120 22.3 89.7d 11.6 + HONO (n=4) NO collection – field DFC systeme 25 ppbv NO (n=4) 120 21.4 40.8 3.1 34 ppbv NO (n=4) 120 21.9 50.4 3.9 56 ppbv NO (n=3) 120 21.2 44.8 6.2 101 ppbv NO (n=4) 120 21.7 36.6 11.7
99.3
90.7
3.5
-70.6 (-69.4)
20.6
99.8 98.7
89.3 89.5
2.4 3.6
-71.1 (-70.0) -70.8 (-69.7)
20.0
99.5
88.2
-3.8
-70.1 (-69.0)
97.5
91.5
2.6
-71.2 (-70.2)
19.5
96.7
89.8
2.8
-71.0 (-69.9)
19.6
103.9 97.6 96.0 97.1
94.3 92.8 92.9 89.5
4.3 -1.8 -2.0 1.0
-72.9 (-71.7) -70.7 (-69.6) -71.5 (-70.4) -71.0 (-69.9)
19.5
Mean Standard error (1 σ)
98.5 3.5
91.7 3.4
1.1 5.1
-71.1 (-70.0) 1.1 (1.1)
19.7 0.8
Time T (min) (°C) NO2 collection – laboratory DFC system 1002 ppbv NO2 (n=4) 135 23.7 Standard error (1 σ) Sample
NO collection – laboratory DFC system 12 ppbv NO (n=3) 120 23.0 34 ppbv NO (n=4) 120 24.8 101 ppbv NO (n=4) 120 23.1
876 877 878 879 880 881 882 883 884 885 886 887 888
a: Out of 56 NO and NO2 tank collection samples, 52 samples yielded consistent results
wherein 4 samples were detected as outliers on the basis of erroneous concentrations. These outliers were not included in this table. b: NO (NO2) recovery was calculated by dividing measured NO2-+NO3- concentration by the theoretical concentration calculated using the collection time, sample flow rate (1.6 slpm), NO (NO2) concentration, and the TEA solution volume. The TEA solution volume was corrected for evaporative loss by weighing the gas washing bottle containing the solution before and after each sample collection. c: Relative to N2 in the air. δ15N values before the isobaric correction are shown in the brackets. d: RH was measured after the HONO scrubber instead of in the Teflon chamber. e: The chamber purging flow rates were 20 slpm, 15 slpm, 9 slpm, and 5 slpm for the 25 ppbv NO, 34 ppbv NO, 56 ppbv NO, and 101 ppbv NO collection, respectively.
ACS Paragon Plus Environment
∆17O (‰)
Page 41 of 44
Environmental Science & Technology
889 890
Figure 1. (a) Schematic of the DFC system (not to scale) consisting of the following: (1)
891
diaphragm pump, (2) air purification columns, (3) drying columns, (4) humidifier, (5)-(7)
892
NO, NO2, and NH3 reference tanks, (8) mass flow controller, (9) flux chamber, (10)
893
temperature and relative humidity sensor, (11) in-line PTFE particulate filter assembly,
894
(12) HONO scrubber, (13) moisture exchanger, (14) reaction tube, (15) gas washing
895
bottle containing TEA solution, (16) O3 generator, (17) NO-NOx-NH3 analyzer; (b)
896
picture showing the field chamber. Specifications of each component of the DFC system
897
are given in the Table S2 in the SI.
ACS Paragon Plus Environment
Environmental Science & Technology
898 899
Figure 2. Illustration of the blank-matching strategy for correcting N blanks associated with the
900
TEA solution and denitrifier method.
ACS Paragon Plus Environment
Page 42 of 44
Page 43 of 44
Environmental Science & Technology
901 902
Figure 3. (a) The measured δ15N of RSIL20 (δ15NRSIL20-m) as a function of the fraction of
903
analyte N2O-N derived from the total N blank (fB). Sample injection volume, standard
904
deviation of δ15NRSIL20-m, and number of replicates for the individual runs are given in the
905
brackets. The dot in red (56 ppbv collection sample) was not included in the linear
906
regression. (b) The measured δ15N of the NO collection sample as a function of the
907
sample NO2-+NO3- concentration. The dash line and the shaded area represent the mean ±
908
(1 σ) of the δ15N of the NO tank measured using the modified EPA NOx collection
909
method.
910 911 912 913 914 915
ACS Paragon Plus Environment
Environmental Science & Technology
916 917
Figure 4. NO emission (lines) and δ15N-NO (dots) results from the laboratory (a) and
918
field (b) rewetting experiments. The error bar on the x-axis denotes the time span of each
919
collection sample. In the field rewetting experiment, four soil plots were respectively
920
wetted on four consecutive days using 500 mL of MilliQ water (black), 20 mM KNO3
921
(red; δ15N=46.5±0.3‰), 10 mM NaNO2 (dark blue; δ15N=1.0±0.4‰), and 20 mM NH4Cl
922
(light blue; δ15N=1.7±0.1‰) solutions.
ACS Paragon Plus Environment
Page 44 of 44