Interactive effects of MnO2, organic matter and pH on abiotic formation of N2O from hydroxylamine in artificial soil mixtures.

Abiotic conversion of the reactive nitrification intermediate hydroxylamine (NH2OH) to nitrous oxide (N2O) is a possible mechanism of N2O formation during nitrification. Previous research has demonstrated that manganese dioxide (MnO2) and organic matter (OM) content of soil as well as soil pH are important control variables of N2O formation in the soil. But until now, their combined effect on abiotic N2O formation from NH2OH has not been quantified. Here, we present results from a full-factorial experiment with artificial soil mixtures at five different levels of pH, MnO2 and OM, respectively, and quantified the interactive effects of the three variables on the NH2OH-to-N2O conversion ratio (RNH2OH-to-N2O). Furthermore, the effect of OM quality on RNH2OH-to-N2O was determined by the addition of four different organic materials with different C/N ratios to the artificial soil mixtures. The experiments revealed a strong interactive effect of soil pH, MnO2 and OM on RNH2OH-to-N2O. In general, increasing MnO2 and decreasing pH increased RNH2OH-to-N2O, while increasing OM content was associated with a decrease in RNH2OH-to-N2O. Organic matter quality also affected RNH2OH-to-N2O. However, this effect was not a function of C/N ratio, but was rather related to differences in the dominating functional groups between the different organic materials.

Nitrous oxide (N2O) is a potent greenhouse gas that can be formed by several soil processes, such as microbial nitrification and denitrification. The N2O production from nitrification, especially from its reactive intermediate hydroxylamine (NH2OH), has received increasing attention in the recent past, fostered by the development of analytical techniques for the determination of the 15N site preference in the N2O molecule that allows for constraining the contribution of different source processes to total N2O formation1234. Also, increasing knowledge from molecular biological and genetic studies has contributed to elucidating the different N2O formation mechanisms during nitrification3. Still, the role of NH2OH in N2O formation in the soil is insufficiently understood. While there is evidence, e.g., from measurements in wastewater treatment systems that NH2OH can contribute about 65% of total N2O formation2, the formation of N2O from NH2OH in soil and its controlling factors have rarely been studied56.

Hydroxylamine was first identified by Lees (1952)7 as an intermediate of the first step of nitrification by n class="Chemical">ammonia oxidizing bacteria (AOB), in which ammonia is oxidized to nitrite. Understanding the nitrification process in ammonia-oxidizing archaea (AOA), however, is much more fragmentary, but NH2OH has been identified as an intermediate of ammonia oxidation also in AOA8. In most circumstances, NH2OH is quickly oxidized to nitrite in the periplasm of the AOB, and N2O may be produced as a side product during this process3. However, also a leakage of NH2OH from the periplasm across the outer membrane of the AOB into the soil matrix, followed by a chemical reaction with soil constituents yielding N2O, could be a potential mechanism of N2O formation during nitrification. This assumption is supported by the fact that AOB can take up NH2OH from the surrounding medium9 as well as by the observation that the medium of AOB cultures contains measurable amounts of NH2OH. The latter was found for Nitrosomonas europaea under oxic conditions, both for wild-type N. europaea and even more so for NirK and NorB-deficient mutants10. In accordance with this assumption, a positive relationship between NH2OH content of the soil and soil N2O emissions under oxic conditions has been detected in natural forest soil samples11. In addition, abiotic formation of N2O from NH2OH has been observed in sterilized soil samples from different ecosystems6.

In soil, N2O can be formed chemically, among a range of possible reactions, according to the following equations12:

Owing to its high oxidization potential, manganese dioxide (MnO2) acts as a strong oxidant in soil that plays an important role not only in the turnover of organic substances1314, but also in the N cycle15, even under anoxic conditions1617. Soil organic matter (SOM) plays a crucial role in the storage and release of N as well as in the emission of N2O from soils. Quick disappearance of nitrite and nitrate within a few hours after addition has been observed in forest soils181920, whereas NH2OH disappeared completely in soil several minutes after addition511. Abiotic reactions of SOM and inorganic N may contribute to the quick disappearance, as nitrite and nitrate can react with SOM or dissolved organic carbon (DOC), leading to the formation of organic N, such as nitroso and nitro compounds2122, while NH2OH can also react with carbonyl groups to form oximes2324:

The quality of SOM, or more specifically the C/N ratio and the type and abundance of functional groups, influence the bonding of inorganic N to SOM22. Phenolic lignin derivatives, an important constituent of SOM, can covalently bind reactive N compounds and thereby stabilize N in soil2526. The N binding form can be affected by the plant species from which the SOM is derived due to the different characteristics of phenolic compounds, e.g. condensed or hydrozable tannin27.

Soil pH is another key factor influencing most nitrogen transformations in soil. High pH can lead to an increase of chemical n class="Chemical">N2O production involving nitrite by favoring nitrite accumulation, either directly through increasing nitrite stability, or indirectly by inhibiting biological nitrite oxidation due to a higher concentration of free NH3 (an inhibitor of nitrite oxidizers) in the soil28. In contrast, high soil N2O emissions have also been observed in acid forest soils2930. In this case, the effect of pH on enzyme activities during denitrification and nitrification was suggested as the main reason31. However, also chemical reactions that produce N2O in the soil, such as the reaction of nitrite with SOM and the reaction of NH2OH with MnO2, are subject to a strong pH dependence and can contribute substantially to N2O emissions under acidic conditions323334.

The aim of this study was to quantify the interactive effects of the major control factors of abiotic n class="Chemical">N2O formation from NH2OH in soil, i.e. MnO2 content, pH and OM quantity and quality, by means of experiments with artificial soil mixtures. We hypothesized that the control factors interact with each other in the following way: At higher pH, unprotonated NH2OH would react more readily with carbonyl groups of OM, leading to oxime formation and making NH2OH less available for oxidation to N2O by MnO2. Lower soil pH would lead to increased protonation of NH2OH, making NH2OH more stable against the reaction with carbonyl groups of OM and more prone to the reaction with MnO2, leading to higher N2O formation from the same amount of NH2OH (Fig. 1). To test these hypotheses, we performed two laboratory experiments with artificial soil mixtures, which were produced from pure quartz sand, quartz powder, kaolin clay, MnO2 powder and different plant-derived organic materials, resembling SOM of different quality, at different mixing ratios. In these experiments, N2O formation was determined after NH2OH addition to the different mixtures at different pH levels and related to the different control factors.

Figure 1

Hypothetical model of NH2OH release by ammonia-oxidizing bacteria to the soil environment and potential abiotic reactions of NH2OH with MnO2 and organic matter in the soil at different pH conditions (R1R2C=O represents carbonyl groups of SOM).

AMO is ammonia monooxygenase; HAO is hydroxylamine oxidoreductase.

Results and Discussion

RNH2OH-to-N2O at different pH, MnO2 and OM contents (%)

In the present study all three factors, i.e. pH, MnO2 and OM content, affected RNH2OH-to-N2O from peat moss significantly (Fig. 2, S1 and S2). The RNH2OH-to-N2O increased greatly with an increase in MnO2 content from 0% to 0.1% (Fig. 2). This finding is consistent with Bremner et al.5, who studied 19 soils with a wide range of properties and found that the formation of N2O by decomposition of NH2OH was highly correlated with oxidized Mn content of the soils. The fact that NH2OH was used in the past for the selective extraction of Mn oxides from soil samples35 indicates that NH2OH can efficiently reduce Mn(IV) to Mn(II) or Mn(III) (and in turn is oxidized to N2O) in natural soil samples. With increasing OM content, RNH2OH-to-N2O decreased remarkably, especially at high pH (Fig. 2c,d,e). For example, an increase in OM by only 1% at 0.01% MnO2 led to about 50% and 80% decrease in N2O emissions at pH 3 and pH 7, respectively (Fig. 2e, S2). This could be caused by the oxime-forming reaction between NH2OH and carbonyl groups of OM, such as in quinones. The oximes may undergo a tautomeric equilibrium with their corresponding nitrosophenol forms23. In fact, NH2OH has been used in a number of previous studies to determine the carbonyl content of humic substances36, indicating a high affinity of NH2OH to OM that contains carbonyl groups. In the absence of OM and MnO2, increasing pH led to a slight increase in RNH2OH-to-N2O due to the self-decomposition of NH2OH at higher pH, whereas in the presence of OM and absence of MnO2 nearly no NH2OH was converted to N2O (Fig. S2a). In contrast, the effect of increasing pH on RNH2OH-to-N2O became negative already in the presence of 0.01% MnO2 (Fig. S2b). This finding suggests that acidic conditions are favorable for the redox reaction between NH2OH and MnO2.

Figure 2

Mean NH2OH-to-N2O conversion ratios (RNH2OH-to-N2O) in artificial soil mixtures at different pH as well as MnO2 and organic matter (OM, peat moss) contents.

The total amount of NH2OH added was 5 nmol (equivalent to 70 μg N per kg dry material). Different symbols represent RNH2OH-to-N2O at different OM content (n = 3, SD < 5%, not shown).

Also strong interactive effects of pH and MnO2, pH and OM, and OM and MnO2 were observed for the conversion of NH2OH to N2O. The largest RNH2OH-to-N2O found in the present experiment was 81.5% in the absence of SOM at pH 3, and with a MnO2 content of 0.1% (Fig. 2a), while the lowest RNH2OH-to-N2O was about 9%, when SOM content was 10% in the presence of 0.1% MnO2 at pH 7 (Fig. 2e). This suggests that even at the highest MnO2 level and in all other respects optimal conditions a small fraction of NH2OH had not been converted to N2O, but to some other unidentified product.

In the treatments without OM, MnO2 had only a small effect on RNH2OH-to-N2O at all pH conditions, while it had a larger effect especially at higher OM content (Fig. 2, S1), suggesting a strong competition between OM and MnO2 for NH2OH. The competition was biased by pH, with lower pH favouring the reaction of NH2OH and MnO2, while higher pH favoured the reaction of NH2OH with OM. These findings confirmed our hypothesis that at low pH NH2OH is more protected against reaction with OM and more available for the oxidation by MnO2 due to the higher degree of NH2OH protonation at lower pH.

RNH2OH-to-N2O as a function of pH, MnO2 content and OM quality

Organic matter quality had a clear influence on RNH2OH-to-N2O in this study (Fig. 3, S3, and S4). Most of the OM types were associated with a significantly lower RNH2OH-to-N2O compared to the mixtures without OM within the pH range of the experiment. In general, the inhibitory effect of the organic materials on the conversion of NH2OH to N2O showed a clear pH dependency, but was not a function of C/N ratio (Fig. 3, S3). At acidic conditions (pH 3–4), peat moss and watermilfoil with their relatively large C/N ratio inhibited RNH2OH-to-N2O the least, while the cyanobacterium material and clover had a stronger inhibitory effect on RNH2OH-to-N2O despite their smaller C/N ratio (Fig. 3a,b). The differences between peat moss, cyanobacterium and watermilfoil material as OM became smaller at higher pH, and were no longer significant at pH 7 in the presence of 0.01% MnO2 (Fig. 3e), while clover showed always the smallest RNH2OH-to-N2O at all pH levels. In the absence of MnO2, all OM forms showed a RNH2OH-to-N2O close to zero, except for the watermilfoil material that was associated with a RNH2OH-to-N2O significantly above zero within the pH range 3–6 (Fig. S4a). A possible explanation could be the fact that, in contrast to the other OM sources, the watermilfoil material contained about 0.03% Mn (Table 1), which could have caused the N2O emission after NH2OH addition even without external MnO2 addition.

Figure 3

Mean NH2OH-to-N2O conversion ratios (RNH2OH-to-N2O) in artificial soils at different pH and MnO2 content, and for organic matter of different origins at a fixed content of 2.5% (w/w).

The total amount of NH2OH added was 5 nmol. Different symbols represent RNH2OH-to-N2O for the artificial soil mixtures with the different organic materials (n = 3, SD < 5%, not shown).

Table 1

Element contents (%) and C/N ratios of the organic materials used in this study.

	C	N	C/N	Al	Ca	Fe	K	Mg	Mn	Na	P	Si
Peat moss	41.3!	0.6	67.2	0.03	0.13	0.06	0.06	0.07	<0.01	0.01	0.03	0.08
Watermilfoil	35.4	2.1	17.0	0.12	2.26	0.11	1.21	0.25	0.031	0.67	0.12	0.21
Clover	41.4	3.7	11.3	<0.01	1.10	0.01	2.68	0.20	<0.01	<0.01	0.34	0.03
Cyanobacterium	44.9	9.9	4.5	0.02	0.31	0.09	1.22	0.31	<0.01	1.36	0.92	0.07

All elements are reported as % of dry weight (mean of three replicates). The standard deviation is 3% for the values larger than 1%, 20% for the values smaller than 0.1%, and 10% for the values in the range of 0.1% to 1%.

We assumed that RNH2OH-to-N2O would be a function of the C/n class="Chemical">N ratio of the different SOM types, as larger C/N ratios would be indicative of a lower degree of N-containing functional groups, i.e. leaving a higher chance for NH2OH to react with SOM and not to be converted to N2O. However, as stated above we did not observe any clear relationship between C/N ratio and RNH2OH-to-N2O, e.g. peat moss had the largest C/N ratio, but did not lead to the lowest RNH2OH-to-N2O. Instead, clover with a much lower C/N ratio had the largest inhibitory effect on RNH2OH-to-N2O. The addition of 2.5% dry clover powder (C/N ratio = 11.3) to the artificial soil mixture decreased RNH2OH-to-N2O by 48% at pH 3 (Fig. 3a), which was similar to the effect of 10% peat moss (C/N ratio = 67.2) at the same pH (Fig. 2a). The reason for this observation could lie in the differences in functional groups between the different organic materials used in this study.

A better insight into the effects of C and N functional groups of the different organic materials was obtained from NMR analysis. The peat moss OM had the lowest proportion of ester or amide carbonyl at around 170 ppm of all materials (Fig. 4, Table 2). This is in accordance with the observation that – despite having the largest C/N ratio – peat moss OM had a lower inhibitory effect on RNH2OH-to-N2O compared to clover and watermilfoil OM (if the background MnO2 effect was subtracted), i.e. the lack of almost any carbonyl groups in peat moss was clearly visible in its chemical behaviour toward NH2OH. In addition, peat moss OM exhibited the largest proportion of O-substituted aliphatic compounds, which might have also contributed to the relatively low inhibitory effect on RNH2OH-to-N2O in comparison to clover and watermilfoil OM. In contrast, cyanobacterium OM had the highest proportion of acid/amide carbonyl of all four organic materials, suggesting the highest inhibitory effect on RNH2OH-to-N2O due to the competitive reaction of carbonyl groups with NH2OH. The clover material, however, contained lower amounts of O-substituted aliphatics and di-O-substituted C in comparison to peat moss and watermilfoil OM, which may have increased its affinity for NH2OH. For the proportion of unsaturated C no clear trend emerged across the different materials, suggesting that the effect of unsaturated C on RNH2OH-to-N2O is of minor importance.

Figure 4

13C- and 15N-CPMAS-NMR spectra of the different organic materials (cyanobacterium, clover, watermilfoil, peat moss) used in the experiment.

Table 2

Relative proportions of chemical features of the different plant materials derived from 13C CPMAS NMR spectra.

Spectral range (ppm)	Chemical features	Found in	Cyanobacterium (%)	Clover (%)	Watermilfoil (%)	Peat moss (%)
45–0	Aliphatic compounds	waxes, suberin, cutin, cyanophycin, chlorophyll (a,b,d)	41	17	15	11
64.5–45	N- and O-substituted aliphats	amino acids, amino sugars, lignin, cyanophycin	19	14	14	12
90–64.5	O-substituted aliphats	polysaccharides, cellulose, hemi-cellulose, starch, pectin, lignin	14	38	42	49
109–90	di-O-substituted C	polysaccharides, cellulose, hemi-cellulose, starch, pectin	3	11	12	14
162–109	unsaturated C, aromatic C	suberin, lignin, chlorophyll	7	11	10	11
190–162	acid, ester, amide	cutin, proteins, cyanophycin, chlorophyll	17	10	7	4

Sums within columns greater than 100 are due to rounding errors.

Development of a stepwise multiple regression model from the artificial soil mixtures and application to natural soils

The multiple regression model obtained from the first experiments was RNH2OH-to-N2O = 45.9–3.1 SOM + 241.1 MnO2 − 4.5 pH, R2 = 0.62 (P < 0.01), which could explain about 62% variation of RNH2OH-to-N2O, and the contributions of pH, Mn and SOM content to the model’s performance were all significant (P < 0.01). It could well explain the observations (Fig. 3) for peat moss, watermilfoil and clover OM (R2 close to 0.8, P < 0.01, Fig. 5). This demonstrated the general applicability of the model for the OM derived from the different plant and cyanobacterium materials, with different N content, aliphatic C content and C/N ratios. In contrast, the model proved to be not appropriate for the artificial soil mixture without any MnO2, indicated by the decreased goodness of the simulation.

Figure 5

Results of the application of the artificial soil regression model for the calculation of NH2OH-to-N2O conversion ratios (RNH2OH-to-N2O) to artificial soil mixtures amended with the different organic materials (n = 22).

The three points for which RNH2OH-to-N2O was determined at pH 3, 4, and 5 without MnO2 addition were excluded from the simulation.

Finally, RNH2OH-to-N2O was simulated with the same regression model for the natural soils described in Heil et al.6. The results showed that the application of the model to natural soils was prn class="Chemical">omising, no matter if it was applied to fumigated or fresh soils (Fig. 6). The simulated RNH2OH-to-N2O explained more than 90% of the observed rates, especially for cropland, grassland and deciduous forest soils. However, the model failed at correctly predicting RNH2OH-to-N2O for the spruce forest soil of Heil et al.6, which could be related to the high SOM and relatively low MnO2 content of the spruce soil as compared to the other soils. This finding suggests that there is a threshold value for the SOM content of 10% above which – and a MnO2 content of 0.01% below which – the model fails to predict the correct RNH2OH-to-N2O values.

Figure 6

Results of the application of the artificial soil regression model for the calculation of NH2OH-to-N2O conversion ratios (RNH2OH-to-N2O) to six natural fresh and chloroform-fumigated soils as reported in Heil et al.6.

Soil pH, MnO2 and SOM content were identified as crucial control variables of RNH2OH-to-N2O, i.e. the conversion ratio of NH2OH to N2O in the artificial soil experiments of this study. Organic matter derived from different plant species and a cyanobacterium also affected RNH2OH-to-N2O due to the differences in composition, type and abundance of functional groups, as more carbonyl C leads to higher reactivity of NH2OH with organic matter, thereby lowering its availability for the oxidation to N2O by MnO2. The multiple regression model of pH, MnO2 and OM developed here could explain about 60% of the variance of RNH2OH-to-N2O in the artificial soil mixtures, and proved also to be promising for the prediction of RNH2OH-to-N2O of chemical N2O production from NH2OH in natural soils, when SOM content was below 10% and Mn content was larger than 0.01%. If these findings can be confirmed for other soils from different ecosystems, this improved understanding of the controls of N2O formation from the reactive nitrification intermediate NH2OH in soils can have large implications for developing appropriate management options, such as adding organic amendments with suitable chemical characteristics, for mitigating N2O emissions from agricultural land, the largest anthropogenic source of N2O to the atmosphere.

Methods

Experimental setup

Two full-factorial artificial soil experiments were conducted. The first experiment comprised three factors (pH, MnO2 and OM content) and five levels of each factor. The scond experiment comprised also three factors (pH, MnO2 and OM quality) with five levels of pH and MnO2, and four different organic materials at the same concentration level (2.5% w/w on a dry weight basis), but of different quality. Each experiment was conducted in triplicate.

Preparation of the artificial soil mixtures

The artificial soil mixtures consisted of 15% (expressed as percentage of dry weight) fine quartz sand (50% of the particles 0.05–0.2 mm), representing the sand fraction, 65% quartz powder (0.002–0.063 mm), represenn class="Chemical">ting the silt fraction, and 20% kaolin clay (≤0.002 mm), representing the clay fraction, mimicking the soil texture of the agricultural Terrestrial Environmental Observatories (TERENO) field site Selhausen37. Freeze-dried, finely ground and sieved (<0.75 mm) peat moss (Sphagnum magellanicum, collected from Dürres Maar, Eifel, Germany) was amended as SOM to the artificial soil mixtures at levels of 0%, 1%, 2.5%, 5%, 10% dry weight, while the relative amount of sand, clay and silt was reduced according to the amount of peat moss added. The water holding capacity (WHC) was determined for each of the artificial soil mixtures. The WHC increased with increasing organic matter (OM) content, and amounted to 29%, 44%, 55%, 76%, and 132% for the five OM contents, respectively. Each of those artificial soil mixtures was amended with MnO2 (Merck, Darmstadt, Germany) at five different levels (0%, 0.01%, 0.025%, 0.05%, 0.1% Mn), then the ingredients were thoroughly homogenized.

Preparation of artificial soil mixtures with different OM qualities

Organic materials with different C/N ratios (Table 1) were derived from two different plant species, i.e. watermilfoil (Myriophyllum spec.) and clover (Trifolium repens), and from a cyanobacterium (Spirulina platensis). Watermilfoil and clover had been collected previously on the campus of Forschungszentrum Jülich (2004 and 2014, respectively), while the cyanobacterium material had been purchased in 2006 (Concept Vitalprodukte, Schwerte, Nordrhein-Westfalen, Germany). The finely ground, freeze-dried and sieved (<0.75 mm) organic material was amended to the inorganic quartz-kaolin mixture as described above at a rate of 2.5% dry weight, while the relative amount of sand, clay and silt was reduced accordingly. Also for this experiment, each of the artificial soil mixtures was amended with MnO2 at five different levels (0%, 0.01%, 0.025%, 0.05%, 0.1% Mn), and again mixed thoroughly to obtain a homogeneous composition.

Addition of NH2OH to the artificial soil mixtures and analysis of the N2O formed

One gram of each artificial soil mixture was weighed into individual 22-mL gas chromatograph (GC) vials. Subsequently, NH2OH in different buffer solutions was added to each vial to obtain a soil water content of 50% WHC, which required addition of varying volumes of buffer solution to the different soil mixtures depending on the OM content, and adaptation of the NH2OH concentration of each of the buffer solutions accordingly. The total amount of NH2OH added to each of the soil mixtures was always 5 nmol (equivalent to 70 μg N per kg dry material). The pH buffer solutions at pH 3, 4, 5 and 6 were prepared with citric acid (0.1 M) and sodium citrate (0.1 M) according to Gomori38, whereas the buffer at pH 7 was prepared with tris(hydroxymethyl)aminomethane and maleate (Tris-maleate buffer). The vials were closed immediately after NH2OH addition. After 10 hours of incubation, the N2O concentration in the headspace of the vials was measured with a GC equipped with an electron capture detector (Clarus 580, PerkinElmer, Rodgau, Germany). Details of the GC setup and analytical conditions have been described previously11.

Calculation of the NH2OH-to-N2O conversion ratio

The NH2OH-to-N2O conversion ratio (RNH2OH-to-N2O, moles N2O-N per mole NH2OH-N, %) was determined according to the following equation:

where c is the background N2O mixing ratio in the headspace of the control with the same amount of n class="Chemical">water instead of NH2OH solution (nL L−1); c is the N2O mixing ratio in the headspace of the sample with NH2OH addition (nL L−1); the factor 2 represents the molar N ratio of N2O and NH2OH; V is the volume of the vial headspace (0.022 L); V is the molar volume of N2O at standard pressure and room temperature (24.465 L mol−1); n is the amount of NH2OH added to the sample vials (5 nmol).

Determination of the basic properties of the organic materials

Three replicates of each organic material were analyzed to determine its basic properties. The C and N content of the difn class="Chemical">ferent organic materials was analyzed by weighing 200–300 μg dry material into tin capsules, followed by combustion at 1080 °C in an elemental analyzer (EuroEA, EuroVector, Milan, Italy) interfaced to an isotope-ratio mass spectrometer (Isoprime, Isoprime Ltd, Stockport, United Kingdom). The C and N content was determined through peak integration of m/z 44 (CO2) and 28 (N2), respectively, and calibrated against elemental standards.

The elemental composition of the organic materials was analyzed by using inductively coupled plasma optical emission spectrn class="Chemical">ometry (ICP-OES) in the central analytical laboratory (ZEA-3) of Forschungszentrum Jülich. Briefly, 100 mg of sample material were mixed with 3 mL HNO3 and 2 mL H2O2, heated in the microwave at 800 W for 30 min. The mixtures were subsequently filled up to 14 mL and diluted 10-fold with deionized water followed by the ICP-OES measurement.

For the determination of characteristic molecule structures and functional groups of the different organic materials used in the experiments, 13C and 15N cross-polarisation magic-angle spinning (CPMAS) nuclear magnetic resonance (NMR) spectra were obtained. 13C CPMAS spectra were obtained on a 7.05 T Varian INOVATM Unity (Varian Inc., Palo Alto, CA, USA) at a 13C resonance frequency of 75.4 MHz. 15N CPMAS spectra were obtained on a 14.09 T Varian NMR system (Varian Inc., Palo Alto, CA, USA) at a 15N resonance frequency of 60.8 MHz. Samples were packed into 6 mm diameter cylindrical zirconia rotors with Vespel® drive tips and spun at 8000 ± 3 Hz in an HX Apex probe. The spectra were collected with a sweep width of 25 kHz and an acquisition time of 20 ms. In preliminary experiments, the optimal contact time and recycle delay for the cross-polarization experiment were determined. A contact time of 1 ms and a 5 s recycle delay time were used for 13C, whereas a contact time of 1 ms and a 1 s recycle delay time were used for 15N. During cross-polarization the 1H radio frequency (RF) field strength was set to 47 kHz for 13C and to 33.7 kHz for 15N, respectively. The 13C and 15N RF field strength was set to 41 and 41.7 kHz, respectively. An ascending ramp of 15 and 12.2 kHz on the 1H-RF field was used for 13C and 15N during contact time to account for inhomogeneities of the Hartmann-Hahn condition, respectively39. Proton decoupling was done using a spinal sequence with a 1H field strength of 50 and 35.6 kHz, a phase of 4.5° and 5.5°, and a pulse length of 12 and 9.5 μs, respectively.

The free induction decays (FID) were recorded with VnmrJ (Version 1.1 RevisionD, Varian Inc., Palo Alto, CA, USA) and processed with Mestre-C (Version 4.9.9.9, Mestrelab Research, Santiago de Compostela, Spain). All FIDs were fourier-transformed with an exponential filter function with a line broadening (LB) of 20 to 50 Hz. Baseline correction was done using the manual baseline correction function of Mestre-C.

The 13C chemical shifts are reported relative to tetramethylsilane (=0 ppm) using adamantane as an external reference. The relative intensities of the regions were determined using the integration routine of the MestRe-C software. The 15N chemical shifts are reported relative to ammonium nitrate (NH4+ = 0 ppm).

Data analysis

The homogeneity of variance was tested with the Bartlett test. One-way analysis of variance (one-way An class="Chemical">NOVA) of the main controlling factors in the two experiments was performed, followed by a Tukey Honest Significant Difference (HSD) test. A stepwise multiple regression model for the NH2OH-to-N2O conversion ratio was developed on the basis of the co-variables pH, MnO2 and SOM content by using the data from the first experiment. In this case, significance was tested with the F test. Linear regression was performed for simulated and measured RNH2OH-to-N2O in artificial and natural soils described in Heil et al.6 and tested for significance. All analyses were performed with the R software package (version 3.1.0, R Development Core Team, 2013)40.

Additional Information

How to cite this article: Liu, S. et al. Interactive effects of MnO2, organic matter and pH on abiotic formation of N2O from hydroxylamine in artificial soil mixtures. Sci. Rep. 6, 39590; doi: 10.1038/srep39590 (2016).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.