The effect of acidic solutions on the determination of the natural abundance of nitrogen isotopes in ammonia

Ammonia (NH₃) is the most important alkaline gas in the atmospheric environment, and plays a key role in the atmospheric nitrogen (N) cycle^[1,2]. However, excess NH₃ in the atmosphere can react with acidic substances such as sulfur dioxide (SO₂), and nitrogen oxides (NO_x) to form ammonium sulfate ((NH₄)₂SO₄), and ammonium nitrate (NH₄NO₃)^[3,4]. These N-containing compounds are important precursors for the formation of fine particulate matter (PM_2.5), which significantly affects air quality, solar radiation, and human health^[5,6]. It can even be transported long distances through the atmosphere, thereby threatening the environmental security of downwind regions^[7]. Therefore, accurate source identification of atmospheric NH₃ is essential for emission mitigation and air-quality management.

Natural abundance of nitrogen isotopes (δ¹⁵N) provides a promising tool for tracing NH₃ sources because different emissions exhibit distinct isotopic signatures, particularly among primary agricultural sources such as fertilizers and livestock waste^[8−10]. Therefore, the δ¹⁵N-based source apportionment method has been successfully applied to quantify the relative contributions of various sources to atmospheric NH₃ and NH_x deposition in different regions^[9,11,12].

Accurate δ¹⁵N determination requires precise methods, particularly during sample collection and measurement. To date, passive collection of atmospheric NH₃ typically relies on acidic adsorbents, such as sulfuric acid, boric acid, phosphoric acid, and citric acid^[13−16]. These solutions convert gaseous NH₃ to dissolved ammonium (NH₄⁺) for subsequent δ¹⁵N analysis. However, previous applications generally focused on quantifying NH₃ concentrations and did not investigate whether absorption solutions or collection efficiencies may introduce isotopic fractionation or be suitable for different kinds of NH₃ source collections, especially those with low concentrations. In fact, the δ¹⁵N values of samples are highly sensitive to the properties and absorption efficiency of the collecting solution^[9,17]. Strong and weak acidic absorbents differ substantially in their protonation capacity and NH₃-NH₄⁺ conversion efficiency, which can directly influence δ¹⁵N measurements^[18]. For example, strong acids such as sulfuric acid provide abundant H⁺ and rapidly convert gaseous NH₃ into stable (NH₄)₂SO₄, ensuring high trapping efficiency and reducing the possibility of isotope fractionation during collection^[19]. In contrast, weak acids such as boric acid supply limited H⁺, which may lead to incomplete NH₃ absorption, lower recovery rates, and a greater likelihood of kinetic isotope effects when the conversion from NH₃ to NH₄⁺. Moreover, due to the influence of emission sources and meteorological conditions, NH₃ concentrations in the atmosphere vary significantly, which may also affect δ¹⁵N determination^[20,21]. At present, no systematic comparison has been conducted to evaluate the suitability of commonly used acidic solutions for δ¹⁵N determination across different NH₃ concentration ranges.

To address this gap, two widely used acidic absorption solutions, sulfuric acid and boric acid, were selected to evaluate their performance in δ¹⁵N determination using isotope reference materials (USGS25, USGS26, IAEA-N1) at different concentrations. Additionally, field sampling was conducted to demonstrate the applicability of the optimized method for characterizing δ¹⁵N-NH₃ signatures from primary agricultural emission sources. The objectives of this study were: (1) to determine the effect of acidic solutions on δ¹⁵N measurements of NH₃; (2) to compare the accuracy and precision of different acidic solutions for determining δ¹⁵N values; and (3) to observe δ¹⁵N values of major NH₃ emission sources using the optimal acidic solutions.

The sponge-trapping method described by Ti et al.^[22] was used to determine NH₃ emission using sulfuric acid and boric acid solutions. Specifically, the NH₃ released from the substrate solution was sequentially absorbed by an acid-treated sponge. A 100 mL solution of (NH₄)₂SO₄ as the NH₄⁺-N substrate was added to a 500 mL incubation bottle. To collect the emitted NH₃, a circular sponge (8.5 cm in diameter, 1 cm thick) impregnated with an acid-absorption solution was attached to the neck of the incubation bottle. The appearance of the culture bottle was described by Ti et al.^[22]. A hole 1.4 cm in diameter was drilled in the bottle cap, and a rubber tube of the same diameter was inserted into it. The small sponge containing the acid-absorption solution was carefully inserted into the tube to prevent ambient air from affecting the experiment. Then, 5 mL of 0.1 mol L⁻¹ sodium hydroxide (NaOH) solution was injected into the bottle through the circular sponge using a 5 mL syringe. The bottle was immediately sealed with the prepared cap. The emitted NH₃ was absorbed by the acid solution within the circular sponge. Using the syringe injection method, rather than removing the sponge to add NaOH directly, ensured that the internal atmosphere of the bottle remained isolated from external air and was not affected by ambient contamination.

After incubation at 25 °C and 95% relative humidity for 24 h, NH₃ emission was terminated by adding 0.8 mL of acid solution to neutralize excess NaOH. The circular sponge used for NH₃ absorption was extracted with 1 mol L⁻¹ KCl solution at 25 °C and 200 r min⁻¹ for 2.5 h, followed by filtration through Whatman 42 filter paper (2.5 μm). The pH of the filtrate was adjusted to approximately 6, and NH₄⁺-N concentrations in both the extract and the residual solution were determined using a continuous-flow analyzer (Smartchem 200S/N1104238, WESTCO, France). The minimum detection limit for the NH₄⁺-N concentrations was 0.03 mg N L⁻¹.

Moreover, the NH₃ recovery rate of this collection system was calculated by measuring the NH₄⁺-N mass in the sponge and in the substrate before and after the reaction. The NH₃ recovery rate was determined using Eq. (1):

where, M₀ indicates the NH₄⁺-N mass (mg N) in the standard solution before reaction. M₁ represents the NH₄⁺-N mass (mg N) measured in the filtrate after sponge extraction, and M₂ means the NH₄⁺-N mass (mg N) in the standard solution after reaction.

Preparation of NH₃ absorbent solution

Standard solutions of 20, 60, and 100 mmol L⁻¹ NH₄⁺N were prepared using (NH₄)₂SO₄ isotope reference materials (USG-S25, USG-S26, IAEA-N1) to simulate the process that NH₃ is converted into NH₄⁺. A volume of 20 μL of standard solutions was sequentially added to 20 mL of sulfuric acid absorption solution (0.1 mol L⁻¹), and boric acid absorption solution (2 wt.%), respectively. The mixed solutions were reacted for 24 h. Then the solutions absorbed by sulfuric acid (0.1 mol L⁻¹) and boric acid (2 wt.%) were adjusted to pH 6.0 using boiled NaOH solution. After dissolution in an equal volume of ultrapure water, the pH was adjusted to 6.0 with a boiled NaOH solution. The (NH₄)₂SO₄ standard solutions were obtained at concentrations of 20, 60, and 100 μmol L⁻¹ (calculated as N).

Before the formal experiments, standard solutions were prepared using (NH₄)₂SO₄ and N isotope reference materials USGS-25, USGS-26, and IAEA-N-1 at a concentration of 20 mg L⁻¹. These solutions were used to verify the systematic recovery rate of NH₃ emission measured in this study, ensuring the accuracy of the experimental data. Each treatment included three parallel replicates.

Natural abundance of isotopes analysis

The δ¹⁵N values in the samples were analyzed using the chemical analysis method described by Liu et al.^[23]. The N-isotopic composition of all samples was analyzed using an isotope mass spectrometer (Isoprime 100, Isoprime, UK). Briefly, NH₄⁺-N in the samples was oxidized to NO₂⁻, which was then converted to N₂O by NH₂OH under strongly acidic conditions. The natural variation in stable N isotopes is generally expressed using the δ¹⁵N value of the heavy isotope. Internationally, N₂ is used as the standard, with its δ¹⁵N defined as 0‰. Isotope ratio values are reported in parts per thousand relative to atmospheric N₂, as shown in Eq. (2):

where, δ¹⁵N_sample represents the δ¹⁵N value of N₂O produced from the sample. (¹⁵N/¹⁴N)_sample and (¹⁵N/¹⁴N)_standard represent the abundance ratios of¹⁵N to¹⁴N in the sample and in the standard, respectively, and (¹⁵N/¹⁴N)_standard was (3,676 ± 8.1) × 10⁻⁶.

Blank correction of mass spectrometry results is performed using the mass balance equation, as shown in Eq. (3):

where, δ¹⁵N_{blank corr} denotes the δ¹⁵N ratio of N₂O in the sample after blank correction, δ¹⁵N_sample represents the δ¹⁵N ratio of the sample measured by mass spectrometry, and δ¹⁵N_blank indicates the δ¹⁵N ratio of the blank. A_sample means the peak area of the sample, and A_blank means the peak area of the blank.

The two N atoms in N₂O molecules produced through chemical conversion originate from NO₂⁻ and NH₂OH, respectively. When NH₂OH reagents are from the same batch, δ¹⁵N-NH₂OH remains constant. Standard curves were established using isotope reference materials of different isotopic abundances to calibrate the system. The δ¹⁵N-N₂O of the generated N₂O can then be used to back-calculate the δ¹⁵N-NH₄⁺ of the substrate^[23,24].

Statistical analysis

Differences in δ¹⁵N values were examined using the one-way analysis of variance, and Duncan's multiple range test. All statistical analyses were performed with SPSS version. 22.0 (IBM Corp., Chicago, IL, USA). The results were drawn using OriginPro 2025b (OriginLab Corporation, Northampton, MA, USA). A p-value of < 0.05 was considered statistically significant throughout the study.

Experiments were conducted using (NH₄)₂SO₄ and N isotope reference materials (USGS-25, USGS-26, IAEA-N1) as reaction substrates to determine the systematic recovery rate of NH₃ in the experimental apparatus. The initial NH₄⁺-N mass of the emission substrate was 2.00 mg. For the (NH₄)₂SO₄ substrate, the NH₄⁺-N mass values from the three replicates were 0.94, 0.95, and 0.83 mg after the reaction, with an average of 0.91 mg, and a coefficient of variation (CV) of 7.58% (Table 1). The NH₄⁺-N mass measured in the filtrates extracted from the sponges containing sulfuric acid absorption solution was 0.99, 0.98, and 1.09 mg, with an average of 1.02 mg and a CV of 5.96%. The slightly higher CV compared with the ideal threshold of 5% was mainly due to minor analytical variation in the third replicate, which showed a lower NH₄⁺-N mass in the reacted (NH₄)₂SO₄ solution, and a higher content in the sponge extract. This deviation was attributed to minor systematic errors in continuous-flow analysis but had little influence on overall recovery efficiency. When the sponge containing the boric acid absorption solution was used, the NH₄⁺-N mass from the three replicates was 0.94, 0.91, and 0.86 mg after the reaction, with an average of 0.90 mg and a coefficient of variation (CV) of 4.51%. The NH₄⁺-N mass measured in the filtrates extracted from the circular sponges was 0.79, 0.78, and 0.75 mg, with an average of 0.78 mg and a CV of 2.72%. The NH₄⁺-N masses in the sponge extract were lower for the boric acid absorption solution than for the sulfuric acid solution.

When NH₃ was collected using the sulfuric acid absorption method, the recovery rate of NH₄⁺-N ranged from 95.98% to 96.88% (mean = 96.43%, CV = 0.47%), while the recovery rate of NH₄⁺-N ranged from 80.47% to 86.48% (mean = 83.90%, CV = 3.68%) for the boric acid absorption method. Results indicated that the sulfuric acid absorption method had high accuracy and reproducibility of the experimental system for NH₃ emission measurements.

For isotopic measurements, the influence of recovery rate on δ¹⁵N accuracy mainly depends on isotope fractionation (ε) during the isolation process. When the system recovery rate is greater than 95%, the resulting deviation in δ¹⁵N (approximately 3‰ under extreme conditions, ε = 20‰) is relatively small and exerts minimal influence on isotope precision^[17]. Therefore, using sulfuric acid absorption to capture NH₃ can effectively preserve δ¹⁵N values, whereas boric acid is less effective at maintaining isotopic integrity. The NH₃ collection apparatus and determination method developed in this study achieved higher recovery rates than those reported by Camargo Valero & Mara^[25] and Zhou et al.^[26]. For example, Zhou et al.^[26] conducted laboratory experiments using sulfuric acid and boric acid as absorption solutions to evaluate NH₃ emission recovery rates under three pH gradients. They indicated that the NH₃ recovery rate for sulfuric acid (89.7%) was higher than that for boric acid (69.8%).

As shown in Table 2, when standard solutions of the N isotope reference materials USGS-25, USGS-26, and IAEA-N1 were used as volatilization substrates, the collection system exhibited consistently high NH₃ recovery rates. For all three standards, the initial NH₄⁺-N mass was set at 2.00 mg. After reaction, the average residual NH₄⁺-N masses for USGS-25, USGS-26, and IAEA-N1 were 0.89 mg (CV = 8.64%), 0.89 mg (CV = 2.83%), and 0.83 mg (CV = 4.89%), respectively. The average NH₄⁺-N masses absorbed by the circular sponge using the sulfuric acid solution were 1.09, 0.99, and 1.09 mg for USGS-25, USGS-26, and IAEA-N1, with CV values of 6.77%, 5.00%, and 4.05%, respectively. The average recovery rates for NH₃ volatilization from USGS-25, USGS-26, and IAEA-N1 were 99.09%, 97.72%, and 96.03%, respectively, with CV values of 0.16%, 1.24%, and 0.17%, respectively. When the boric acid was used as an NH₃ absorption solution, the average residual NH₄⁺-N masses after reaction for USGS-25, USGS-26, and IAEA-N1 were 0.90 mg (CV = 4.61%), 0.82 mg (CV = 2.15%), and 0.83 mg (CV = 2.02%), respectively. The average NH₄⁺-N masses absorbed by the circular sponge were 0.97, 0.94, and 0.93 mg for USGS-25, USGS-26, and IAEA-N1, with CV values of 0.55%, 2.35%, and 0.45%, respectively. The average recovery rates for NH₃ volatilization from USGS-25, USGS-26, and IAEA-N1 were 93.41%, 90.27%, and 88.45%, respectively, with CV values of 1.93%, 2.20%, and 0.71%, respectively.

These results demonstrate that the system maintained stable, high recovery rates across all N-isotope reference solutions, comparable to those obtained with (NH₄)₂SO₄. The sulfuric acid absorption method also showed higher accuracy and reproducibility in the experimental system for NH₃ emission measurements than the boric acid solution method. The low CVs further confirm the high precision and reliability of the collection system for measuring NH₃ volatilization and its natural isotopic abundance.

Effect of different acid solutions on δ¹⁵N measurement

The δ¹⁵N-NH₄⁺ vs δ¹⁵N-N₂O conversion curves were established for different NH₄⁺ concentrations absorbed by various acidic solutions. As shown in Fig. 1, δ¹⁵N-NH₄⁺ and δ¹⁵N-N₂O exhibit excellent linear relationships (R² = 1) for sulfuric acid absorption solution across different NH₄⁺ concentrations (20, 60, and 100 μmol L⁻¹). Mass balance equations were applied to correct δ¹⁵N-N₂O measurements derived from the conversion of different acidic absorption solutions and different concentrations of (NH₄)₂SO₄ solutions. Before correction, the slope of the δ¹⁵N-NH₄⁺ vs δ¹⁵N-N₂O conversion curves for sulfuric acid solution was 0.45 at 20 μmol L⁻¹, 0.45 at 60 μmol L⁻¹, and 0.47 at 100 μmol L⁻¹, respectively. After correction, the slope of the δ¹⁵N-NH₄⁺ versus δ¹⁵N-N₂O conversion curves was 0.46 at 20 μmol L⁻¹, 0.46 at 60 μmol L⁻¹, and 0.48 at 100 μmol L⁻¹, respectively. The slight increase in the corrected slope indicates that after correction, the slopes of the δ¹⁵N-NH₄⁺ vs δ¹⁵N-N₂O conversion curves approached the theoretical value of 0.5, improving the accuracy of isotope conversion. Besides, the minor variation in slope before and after correction suggests that NH₄⁺ concentration has a weak influence on the isotope conversion relationship, so that sulfuric acid absorption solution exhibits good stability and consistency.

Figure 1. 

Conversion curves of δ¹⁵N-N₂O and δ¹⁵N-NH₄⁺ at different NH₄⁺ concentrations in sulfuric acid (0.1 mol L⁻¹) solution.

As shown in Fig. 2, δ¹⁵N-NH₄⁺ and δ¹⁵N-N₂O exhibit distinct linear relationships (R² = 1) for boric acid absorption solution across different NH₄⁺ concentrations (20, 60, and 100 μmol L⁻¹). Before correction, the slope of the δ¹⁵N-NH₄⁺ vs δ¹⁵N-N₂O conversion curves for boric acid solution was 0.37 at 20 μmol L⁻¹, 0.40 at 60 μmol L⁻¹, and 0.44 at 100 μmol L⁻¹, respectively. After correction, the slope of the δ¹⁵N-NH₄⁺ vs δ¹⁵N-N₂O conversion curves was 0.39 at 20 μmol L⁻¹, 0.47 at 60 μmol L⁻¹, and 0.48 at 100 μmol L⁻¹, respectively. The significant increase in the corrected slope indicates that the slopes of the δ¹⁵N-NH₄⁺ vs δ¹⁵N-N₂O conversion curves approached the theoretical value of 0.5 after correction, especially at high concentrations of 60 and 100 μmol L⁻¹. However, the slope for the low-concentration solution (20 μmol L⁻¹) remains low, indicating that the isotope conversion efficiency of the boric acid absorption system is relatively weak at low NH₄⁺ concentrations.

Figure 2. 

Conversion curves of δ¹⁵N-N₂O and δ¹⁵N-NH₄⁺ at different NH₄⁺ concentrations in boric acid (2 wt.%) solution.

In this study, the NH₄⁺-N concentration influenced the relationship between δ¹⁵N-NH₄⁺ and δ¹⁵N-N₂O. When the concentrations of (NH₄)₂SO₄ standard solutions in acidic absorbents ranged from 20 to 100 μmol L⁻¹, the slopes of the δ¹⁵N-NH₄⁺ vs δ¹⁵N-N₂O conversion curves increased with concentration, both before and after correction, gradually approaching the theoretical value of 0.5 (Fig. 3). Thus, the NH₄⁺ concentration significantly affects the δ¹⁵N value of volatilized NH₃. Similarly, previous studies observed that δ¹⁵N-NH₄⁺ generally decreases slightly with increasing NH₄⁺-N concentration, attributed to rapid NH₄⁺-N accumulation and¹⁵N depletion^[27−29].

Figure 3. 

Slope of conversion curves of δ¹⁵N-N₂O and δ¹⁵N-NH₄⁺ at different NH₄⁺ concentrations in sulfuric acid and boric acid solutions.

Numerous studies have investigated the use of acidic solutions for atmospheric NH₃ absorption, and δ¹⁵N values from different emission sources have also been widely examined^[15,22]. However, few studies have addressed the applicability of acidic solutions for δ¹⁵N measurement across different NH₄⁺-N concentrations. In this study, the suitability of δ¹⁵N measurements of NH₃ absorbed by various acidic solutions was analyzed. When the NH₄⁺-N concentration in acidic solutions is identical, the slope of the relationship curve between δ¹⁵N-NH₄⁺ and δ¹⁵N-N₂O before calibration is closest to the theoretical value of 0.5 for sulfuric acid absorption solution (Fig. 3). Compared to boric acid, sulfuric acid remains suitable for detecting δ¹⁵N-NH₄⁺ even at low NH₄⁺-N concentrations in samples. The conversion curve of δ¹⁵N-NH₄⁺ in the sulfuric acid solution exhibited minimal influence from sample NH₄⁺-N concentration before and after blank correction. Previous studies indicated that low concentrations of NH₄⁺-N increased the proportion of blank contributions, leading to either underestimated or overestimated δ¹⁵N values and resulting in poor reproducibility^[30]. This explains the fact that when higher concentration standards are used in the conversion test, the resulting slope approaches the theoretical value of 0.5 more closely. The results indicated that sulfuric acid was more suitable for δ¹⁵N-NH₄⁺ determination when the NH₄⁺-N concentration range in the sample is wide.

During isotope determination, systematic deviations arise from factors such as the absorber, conversion chemistry steps, sample quantity, reagent purity, and blank contributions^[9]. Therefore, the slope of the relationship curve between δ¹⁵N-NH₄⁺ and δ¹⁵N-N₂O was closer to 0.5 after black correction. Due to the simplicity of preparing sulfuric acid absorption solutions and the lower impurity content in sulfuric acid solutions compared with boric acid and citric acid solids, the impact on δ¹⁵N-NH₄⁺ mass spectrometry results was minimized^[23]. Studies have also shown that acidic substances can influence NH₃ volatilization by regulating pH, with increased pH elevating NH₃ volatilization potential^[31]. In acidic conditions, NH₄⁺ remains predominantly in its ionic form, whereas in alkaline conditions, most NH₄⁺ forms volatile NH₃^[32]. A previous study also observed that higher pH, along with a higher N fractionation factor, occurred during the NH₃ volatilization process^[33]. Sulfuric acid, being a strong inorganic acid, may exhibit greater NH₄⁺ fixation capacity than weaker inorganic acids like boric acid or organic acids like citric acid during absorption and pH adjustment to 6.0^[18]. This reduces the conversion of NH₄⁺ to NH₃ and minimizes isotopic fractionation of both NH₄⁺ and NH₃. Consequently, 0.1 mol L⁻¹ sulfuric acid yields the most reliable results in δ¹⁵N-NH₄⁺ measurements.

Using the 100 μmol L⁻¹ (NH₄)₂SO₄ standard solutions as an example, the accuracy and precision of δ¹⁵N measurements obtained with different acidic solutions were evaluated, and the results are presented in Table 3. For the USGS-25 and USGS-26 standards, the measured δ¹⁵N-NH₄⁺ values using sulfuric acid and boric acid were all in close agreement with their certified reference values of −30.43‰ and 53.75‰, respectively. The deviations between measured and reference values were within ±0.5‰, indicating stable analytical performance. For the IAEA-N1, which has a relatively low δ¹⁵N-NH₄⁺ (reference value 0.41‰). The δ¹⁵N-NH₄⁺ values obtained with sulfuric acid and boric acid were 0.21‰ and 0.27‰, respectively. Although these results showed slightly greater relative variation, they remained within the reference uncertainty range and met analytical requirements.

Overall, the recovery rate of the boric acid solution was relatively low, which could lead to deviations in δ¹⁵N values, particularly at low NH₄⁺ concentrations. In contrast, 0.1 mol L⁻¹ sulfuric acid provided consistent results across all standards. Considering its simple preparation and stability, sulfuric acid is recommended as the optimal absorbent for δ¹⁵N determination, especially for samples with low or fluctuating NH₄⁺ concentrations.

Case study of measurements for δ¹⁵N values of major NH₃ emission sources

Since the sulfuric acid absorption solution provided excellent results for δ¹⁵N determination of the two acid absorption solutions, it was used to conduct NH₃ absorption and δ¹⁵N measurement of primary NH₃ emission sources, including cropland, livestock, orchards, and vegetables. The δ¹⁵N-NH₃ values for orchard and vegetable sources were obtained from Bai et al.^[34], whose sampling setup and acidic absorption solution were identical to those in this study, thereby enabling a direct and valid comparison. As shown in Fig. 4, the mean δ¹⁵N values of NH₃ emissions were −32.87‰ ± 7.41‰ (range: −44.79‰ to −21.56‰) for cropland, −36.64‰ ± 9.19‰ (range: −50.09‰ to −25.47‰) for livestock, −19.63‰ ± 5.09‰ (range: −27.98‰ to −13.29‰) for orchards, and −24.95‰ ± 3.29‰ (range: −29.26‰ to −18.52‰) for vegetables. The δ¹⁵N-NH₃ values for cropland and livestock were significantly lower than those for orchards and vegetables. Our obtained δ¹⁵N values were within the range reported for agricultural sources in previous studies^[24,35,36], which were obtained using passive adsorption samplers, supporting the effectiveness of the sampling results. However, the δ¹⁵N values obtained with the passive sampler in this study were significantly lower than those measured with active samplers, due to isotope fractionation effects between different samplers^[9]. For example, the δ¹⁵N values for livestock measured using active sampling methods^[15] were significantly higher than those obtained using passive samplers^[24,35,36]. Previous studies have shown that δ¹⁵N measurements from active and passive NH₃ samplers can differ by roughly 15‰, reflecting sampler-induced fractionation associated with diffusion limitations and incomplete absorption^[9].

Figure 4. 

δ¹⁵N values for major NH₃ emission sources in sulfuric acid solution. The δ¹⁵N values for cropland and livestock were measured in this study, and for orchards and vegetables were measured in the study of Bai et al.^[34].

NH₃ emission is affected by several factors, including fertilizer application rates, NH₄⁺ substrate concentrations, and temperature^[37,38]. These factors directly or indirectly cause significant variations in the δ¹⁵N values of emitted NH₃^[22,39]. Soil NH₄⁺ and NH₃ volatilization rates exhibited a negative correlation with δ¹⁵N values^[40]. In general, molecules containing¹⁴N react faster than those containing ¹⁵N because the energy required to break or form chemical bonds involving ¹⁵N is greater than that for bonds involving ¹⁴N^[41]. However, partial overlap in δ¹⁵N values among different emission sources may be influenced by variations in geography, season, and temperature^[42]. Future research requires additional high-temporal-resolution field observation experiments to supplement δ¹⁵N values.

Therefore, the absorption and processing method for sulfuric acid solutions is highly suitable for determining the natural isotopic abundance of atmospheric NH₃ and its major emission sources in the field, which contributes to the source apportionment of atmospheric NH₃. In addition, NH₃ emitted from different emission sources exhibit unique δ¹⁵N values, which have been successfully applied to explore N sources, transformation processes, and ultimate destinations under various scenarios^[43−45]. δ¹⁵N-NH₃ values serve as a crucial tool for clarifying atmospheric NH₃ origins and provide precise data support for scientific and efficient fertilization in agricultural fields, as well as atmospheric haze management^[46−48].

This study conducted a systematic investigation of recovery performance, isotope measurement stability, and field observations of NH₃ using sulfuric acid and boric acid absorption solutions by the sponge absorption method. Results indicated that sulfuric acid absorption solution achieved a significantly higher overall NH₃ recovery rate than the boric acid absorption solution, and exhibited excellent systematic stability for δ¹⁵N-NH₄⁺ measurement when the NH₄⁺-N concentration range in the sample is from 20 to 100 μmol L⁻¹. Therefore, sulfuric acid is identified as the optimal absorption for accurate δ¹⁵N analysis across a wide range of NH₃ concentrations. This study also successfully applied the method to determine δ¹⁵N-NH₃ values for cropland, livestock, orchards, and vegetable emission sources. Therefore, the absorption method for sulfuric acid solutions is highly suitable for determining NH₃ emissions and δ¹⁵N-NH₃ values of major emission sources in the field, thereby contributing to the source apportionment of atmospheric NH₃.