Formation of simple nitrogen hydrides NH and NH₂ at cryogenic temperatures through N + NH₃ → NH + NH₂ reaction: dark cloud chemistry of nitrogen

Abstract

Although NH₃ molecules interacting with ground state nitrogen atoms N(⁴S) seem not to be a very reactive system without providing additional energy to initiate the chemical process, we show through this study that, in the solid phase, at very low temperature, NH₃ + N(⁴S) reaction leads to the formation of the amidogen radical NH₂. Such a dissociation reaction previously thought to occur exclusively through UV photon or energetic particle irradiation is in this work readily occurring just by stimulating the mobility of N(⁴S)-atoms in the 3–10 K temperature range in the solid sample. The N(⁴S)–N(⁴S) recombination may be the source of metastable molecular nitrogen N₂(A), a reactive species which might trigger the NH₃ dissociation or react with ground state nitrogen atoms N(⁴S) to form excited nitrogen atoms N(⁴P/²D) through energy transfer processes. Based on our obtained results, it is possible to propose reaction pathways to explain the NH₂ radical formation which is the first step in the activation of stable species such as NH₃, a chemical induction process that, in addition to playing an important role in the origin of molecular complexity in interstellar space, is known to require external energy supplies to occur in the gas phase.

---

I. Introduction

Ammonia is one of the most important simple molecules that may be the source of the N–H and N–N bonds that lead to the formation of pre-biotic molecules in interstellar space and in some planetary and satellite atmospheres. As a consequence of its high microwave absorbance, several studies consider ammonia molecules to be the primary source of opacity in many extraterrestrial atmospheres.¹ As a result, numerous space observations and laboratory simulations have been carried out to provide a clear description of the role of NH₃ molecules in the chemical activities in space. Hofstadter et al.¹ underlined the importance of latitudinal variations of ammonia in the atmosphere of Uranus, through an analysis of a radio map of the planet made with the very large array, while Loeffler et al. carried out laboratory simulation by heating water-ammonia ices² previously irradiated with protons to simulate Saturn's energetic ion environment. They illustrated that the eruption of high-pressure bubbles of hydrogen and nitrogen molecules might be due to the radiolytic decomposition of water-ammonia ices, one of the phenomena responsible for the variations of NH₃ abundances. Regarding the variations of NH₃ abundances, studies of the ammonia abundances vs. temperatures and photo-dissociation processes in some nearby galaxies, undertaken very recently by Takano et al., have already shown that galaxies with low temperatures tend to have low abundances of ammonia.³ However, those studies have only proved that the low abundances of ammonia cannot be explained only by the photo-dissociation effects and that other chemical processes might be involved in the evolution of their abundances. On the other hand, interstellar ammonia, considered to be originating from molecular cloud materials,⁴ might be one of the formation sources of N–H-bearing reactive species such as NH, NH₂, and N₂H₃. However since the detection of the first interstellar neutral N-bearing molecules,⁵ the ratios of the three simple nitrogen hydrides NH : NH₂ : NH₃ were found to be strongly dependent on dark cloud conditions. In 1991 Millar et al. predicted⁶ a dark cloud value NH₃/NH₂ < 3 abundances, incompatible with the observations of van Dishoeck et al.⁷ or Hily-Blant et al.⁸. In fact, Hily-Blant et al. reported the ratio of NH : NH₂ : NH₃ equal to 5 : 1 : 300 in the cold envelope of IRAS16293-2422, while, using the Caltech Submillimeter Observatory to probe the NH₂ radical in interstellar clouds, van Dishoeck et al. found a value of 0.5 for the NH₂/NH₃ ratio. Those disparities in NH₂ and NH₃ abundance measurements have caused many groups to state that the NH₂ radical may be formed through different reaction mechanisms depending on the dark cloud conditions.

It is well established that the photo-dissociation of NH₃ leads mainly to the formation of NH and NH₂ radicals as primary photoproducts. Consequently, many laboratory investigations have focused their NH₂ formation research on ultraviolet processing of interstellar or satellite containing NH₃ ice analogs^9,10 in order to mimic the photo-chemistry processes of icy objects exposed to UV radiation in different regions of space, such as interstellar clouds, Saturn's rings, comets and planetary atmospheres. In the same context, earlier studies carried out by Miligan et al. and Schnepp et al. have already characterized the formation of NH₂ radicals^11,12 through studies of photolysis of solid ammonia formed at 14 K, where the NH₂ infrared fundamental vibration has been measured at 1499 cm⁻¹. In those studies related to ammonia decomposition, the NH₂ radical has been characterized only by the bending mode signal around 1500 cm⁻¹ as the NH stretching is unfortunately partially obscured by the large IR-absorption bands of ammonia aggregates. Similarly, in 1984, Nishi et al. underlined¹³ the formation of a hydrogen bound NH₂–NH₃ complex on the surface of irradiated icy solids containing NH₃. More recently, Loeffler et al. studied, using IR spectroscopy and mass spectrometry, the UV photolysis of solid ammonia and ammonia-dihydrated samples formed at 40 K.¹⁴ They showed that photolysis of NH₃ leads to the formation of NH₂ as a first primary photoproduct. In this context we have also investigated the UV photolysis of ammonia¹⁵ ice at 3 K, confirming the results obtained by Loeffler et al. and showing that the only detectable IR-signal due to the photo-dissociation of solid NH₃, a wide absorption band due to the NH₂ radical, is located around 1500 cm⁻¹ (Fig. 1).

Fig. 1 UV-irradiation of ammonia ice formed at 3 K. (a) Before irradiation and (b) after 30 min irradiation using a UV-hydrogen lamp.

This band has been attributed to the NH₂ radical interacting with ammonia aggregates by carrying out the same experiments using neon matrix isolation.¹⁵Fig. 2 shows the formation of different complexes between NH₂ and NH₃ while the ammonia concentration increases in the neon matrix. The formation of (NH₂)(NH₃), (NH₂)(NH₃)₂ and (NH₂)(NH₃)_n was clearly established thanks to the comparison between the theoretical and experimental vibrational frequencies.

Fig. 2 Formation of NH₂ and (NH₂)(NH₃)_n species during the photolysis of NH₃/Ne mixtures at 3 K. NH₃ concentration effects. (a) NH₃ : Ne = 0.002 : 100, (b) NH₃ : Ne = 0.008 : 100, (c) NH₃ : Ne = 0.004 : 100, (d) NH₃ : Ne = 0.1 : 100, and (e) NH₃ : Ne = 0.2 : 100.

Although from the previous studies NH₃ photo-dissociation might be considered an important source of NH and NH₂ radicals, the aim of the present study is to characterize the formation of reactive nitrogen hydride radicals from non-energetic processes involving non-reactive but very abundant species in space, namely ground state nitrogen atoms N(⁴S) and ammonia molecules NH₃. We show through this experimental investigation that while the absence of additional energy sources associated with an extremely low temperature may present conditions less favorable for the N(⁴S) + NH₃ reaction to occur, NH and NH₂ radicals may be easily formed in the solid phase just by stimulating a mobility of N-atoms around NH₃ molecules in the 3–10 K temperature range. These results show that complex nitrogen chemistry, involving nitrogen atoms and initiated by thermal processes, may take place in dense molecular clouds. Indeed, nitrogen is considered as one of the most abundant elements in various very cold regions of space.¹⁶ Knauth et al.¹⁷ detected N₂ molecules in interstellar space, with a column density of 4.6 × 10¹³ cm⁻², several orders of magnitude lower than that of atomic nitrogen 2.0 × 10¹⁷ cm⁻², demonstrating that nitrogen is mainly atomic. On the other hand, several interstellar gas-grain chemistry models predicted that N₂ should be formed on ice mantles through N–N addition reactions and that much of the missing nitrogen is present in icy grains. In such a context as for N + N recombination on ice mantles, NH₃ + N reaction might be also very frequent. Paradoxically, to date, no theoretical investigations have even been carried out to describe the reaction between ground or excited state nitrogen atoms and NH₃ molecules. Based on previous theoretical models describing reactions with N atoms, many studies have already shown that all reactions involving excited N-atoms such as N(²D) and N(⁴P) are highly exothermic, with almost no activation energy, while reactions involving ground state nitrogen atoms N(⁴S) are found to be endothermic and they need a high additional energy supply to the proceed. Even with the lack of N + NH₃ theoretical models, we show, in the present article, that the hydrogen atom abstraction reaction from the NH₃ molecule is due to the mobility of nitrogen atoms in the solid phase. This is the first study of the NH₃ + N(⁴S) reaction which was carried out in the solid phase at very low temperature. We have explored such a reaction to characterize the three simple nitrogen hydrides NH : NH₂ : NH₃ through the N + NH₃ → NH + NH₂ reaction pathway. We show, in the course of this study, that the NH₂ radical is formed from ammonia under low temperatures, low pressures, and non-energetic conditions, circumstances different from those involving high energy photons or particles. As we are particularly interested in solid-phase reactivity, we conducted a detailed investigation of the NH₃ + N(⁴S) reaction by co-condensing NH₃ molecules and N atoms at 3 and 10 K. The obtained solid samples were further analyzed spectroscopically using a Fourier Transform Infrared (FT-IR) spectrometer.

II. Technical background

The experimental methods used for the present study have been previously described.^18,19 Nevertheless, it is necessary to remind the reader of the basic scheme. The setup is maintained at 10⁻⁸ mbar, and samples were prepared by condensing gaseous mixtures on the surface of a substrate mirror maintained at 3 K. The co-deposition lasts for 30 min. To this end, a pulsed tube, closed cyclic helium cryogenerator is used (Cryomech PT405, USA). Prior to condensation onto the cold mirror, nitrogen atoms are obtained by passing pure nitrogen gas through a microwave-driven radical atomic source (SPECS PCR-ECS). Nitrogen atoms are generated with a flux of about 10¹⁵ atoms cm⁻² s⁻¹, corresponding to a 4% N₂-dissociation rate. The dissociation rate has been evaluated by monitoring the IR signals corresponding to solid N₂ formed at 3 K under the same experimental conditions with the microwave discharge on and off. The obtained samples are analyzed using a Bruker 120 HR Fourier Transform InfraRed (FT-IR) spectrometer in the transmission-reflection mode between 4500 and 500 cm⁻¹. The angle of the IR beam is 8° with respect to the normal of the deposition mirror. A resolution of 0.5 cm⁻¹ is used and 300 scans are recorded for each spectrum. Pure NH₃ molecules are concomitantly injected, at 3 or 10 K, with the N/N₂ discharged mixture and for the sake of comparison all the IR spectra are registered at 3 K. The formed solid sample is kept in a chamber under vacuum (10⁻⁸ mbar) where the only pollutions detected as traces are CO, CO₂ and H₂O, during the sample deposition when the pressure increased to 10⁻⁵ mbar. Molecular N₂ and NH₃ were purchased from “Air Liquide” and Messer, respectively, with a purity of 99.9995%. The purity of samples was confirmed spectroscopically. Experiments at 3 and 10 K were carried out, however, the 3 K lowest temperature was preferable for the present study in order to control the mobility of nitrogen atoms trapped in the solid samples. In fact, at temperatures as low as 3 K, the resulting samples may be warmed up from 3 to 10 K to study the diffusion of nitrogen atoms and then the evolution of the reaction, thermally stimulated.

III. Results

Fig. 3a shows the spectrum of a sample obtained by co-injection of ammonia and N/N₂ reactants at 3 K, corresponding to NH₃ + N + N₂ reaction. As a reference, a spectrum of the sample containing only NH₃ and N₂ as reactants, at the same temperature, is shown in Fig. 3b. Comparison between these two spectra reveals that at 3 K, there are no new reaction products due to the NH₃ + N reaction.

Fig. 3 NH₃ and NH₂ spectral regions. (a) NH₃ + N + N₂ reaction and co-injection of the NH₃ and N/N₂ mixture at 3 K. (b) NH₃ + N₂ reaction and co-injection of NH₃ and N₂ gas at 3 K.

In fact by zooming in the NH₂ characteristic spectral region around 1500 cm⁻¹, Fig. 4 shows no signal due to NH₂ species which, as mentioned above, has been measured in the solid phase by Loeffler et al.¹⁴ at 1508 cm⁻¹. As NH₃ + N + N₂ and NH₃ + N₂ systems give the same IR response, we attest that no reaction may be possible between ammonia and nitrogen atoms at 3 K. These first results also prove that the N-atoms reaching the solid sample are mainly ground state nitrogen atoms N(⁴S), atomic species less reactive than excited nitrogen atoms N(²D) and N(⁴P). In order to monitor the NH₃ + N reaction, the former solid sample obtained at 3 K has been kept in the dark, far away from any excitation source such as room light or IR radiation from our spectrometer, for a period ranging from 1 min to several hours. This was performed to ensure that all the reactants are in their ground states and that the N + NH₃ reaction is not triggered by any other external source of energy. After several hours in the dark, the recorded IR spectrum of the sample is similar to that obtained just after N + NH₃ co-deposition, showing that our sample was not altered in any way while kept in the dark at 3 K. In fact, at 3 K, the mobility of the reactants, particularly nitrogen atoms, is very limited and thus reduces the encounter probability and hence the interaction between the reactants. In order to induce some motion, the sample was gradually heated from 3 to 10 K and an infrared spectrum was subsequently acquired at 3 K.

Fig. 4 NH₂ spectral region. (a) NH₃ + N + N₂ reaction and co-injection of the NH₃ and N/N₂ mixture at 3 K. (b) NH₃ + N₂ reaction and co-injection of NH₃ and N₂ gas at 3 K.

Fig. 5a shows the results of the sample heating between 3 and 10 K. Detectable signals due to the NH₂ radical around 1500 cm⁻¹ are observed, but only in samples containing nitrogen atoms and NH₃ molecules. Fig. 5b shows the results of the heating of our reference sample containing just NH₃ and N₂ as reactants and no new signals are identified.

Fig. 5 NH₂ spectral region. (a) NH₃ + N + N₂ reaction, co-injection of the NH₃ and N/N₂ mixture at 3 K and heating of the sample at 10 K. (b) NH₃ + N₂ reaction and co-injection of NH₃ and N₂ gas at 3 K, and heating of the sample at 10 K. All the IR spectra are recorded at 3 K.

Similar experiments have been carried out by varying the amount of injected NH₃ during the NH₃ + N/N₂ co-injection. Fig. 6, corresponding to the NH₃ spectral region, and Fig. 7, to that of the NH₂ radical, show the influence of ammonia concentration on NH₃ + N reaction. The spectra presented in these two figures result from a reactant co-deposition at 3 K, followed by a sample heating at 10 K to thermally induce the NH₃ + N reaction. In order to have an order of magnitude of the ammonia concentrations in each studied sample, we have calculated the column density n (molecules per cm²) of ammonia, using the band area of the absorption ν₂ vibrational mode, associated with its appropriate²⁰ band strength (A = 1.2 × 10⁻¹⁷ cm per molecules). The column densities are defined as:²¹

with the integral being the band area (cm⁻¹) of the absorption vibration mode of the considered chemical species. The correction factor cos(8°) takes into account the angle of the IR beam with respect to the normal of the deposition mirror, and the division by two corrects the fact that our samples are analyzed twice in the transmission-reflection mode.

Fig. 6 NH₃ spectral region. NH₃ + N + N₂ reaction, co-injection of the NH₃ and N/N₂ mixture at 3 K and heating of the sample at 10 K. (a) [NH₃] = 0.2 × 10¹⁷ molecules per cm², (b) [NH₃] = 0.7 × 10¹⁷ molecules per cm², and (c) [NH₃] = 2.5 × 10¹⁷ molecules per cm². All the IR spectra are registered at 3 K.

Fig. 7 NH₂ spectral region. NH₃ + N + N₂ reaction, co-injection of the NH₃ and N/N₂ mixture at 3 K and heating of the sample at 10 K. (a) [NH₃] = 0.2 × 10¹⁷ molecules per cm², (b) [NH₃] = 0.7 × 10¹⁷ molecules per cm², and (c) [NH₃] = 2.5 × 10¹⁷ molecules per cm². All the IR spectra are registered at 3 K.

The association of Fig. 6 and 7 allows for the assignment of the detected signal at 1508 cm⁻¹ to the NH₂ radical. The attribution of this signal, observed even at very low NH₃ concentration (Fig. 6a and 7b), is in good agreement with Loeffler's measurements.¹⁴ The signals detected at 1480 and 1520 cm⁻¹, observed at relatively high NH₃ concentrations, might be due to the NH₂ radical interacting with NH₃ molecules (Fig. 6b, c and 7b, c). Our assignments are also consistent with our previous recent study characterizing complexes between NH₂ and NH₃ in a neon matrix through the UV photolysis of ammonia.¹⁵ We showed, by combining neon matrix isolation studies with theoretical calculations, that trapped in the neon matrix the NH₂ radical and (NH₂)(NH₃)_n aggregates show characteristic signals at 1501.1 and 1487.1 cm⁻¹, respectively. Table 1 gathers the calculated and measured values of the spectral positions of NH₂ and complexes between NH₂ and NH₃.

Table 1 Calculated and measured spectral positions (cm⁻¹) of NH₂ and complexes between NH₂ and NH₃

	The present study	Neon matrix observation^a	Calculated^a
a Ref. 15.
NH2	1508	1501.1	1505.6
(NH2)(NH3)	—	1516.5	1518.9
(NH2)(NH3)2	—	1529.6	1530.6
Larger (NH2)(NH3)n aggregates	1482	1487.1	—

---

In our previous study, the formation and isolation in the neon matrix of NH₂, (NH₂)(NH₃), (NH₂)(NH₃)₂ and (NH₂)(NH₃)_n species were clearly established by irradiating samples with different NH₃/Ne amounts. We show in the present study that the NH₂ radical is formed by stimulating thermally the mobility of N(⁴S)-atoms around ammonia molecules. However from Fig. 6 and 7, we note that while in diluted samples ([NH₃] = 0.2 × 10¹⁷ molecules per cm²) the NH₂ signal is observed as the main reaction product, an increase in the NH₃ concentration stimulates directly the formation of larger (NH₂)(NH₃)_n aggregates. The direct formation of (NH₂)(NH₃)_n under our experimental conditions may be due to the fact that the NH₃ + N reaction takes place only by heating the sample from 3 to 10 K which may also favor the formation of ammonia aggregates in the nitrogen matrix.

One of the advantages of carrying out this experiment study at 3 K, in addition to proving that all chemical processes start from reactants in their ground states and that the reactions occur without providing any external energy to the NH₃ + H system, is the possibility to deduce the amount of reacting ammonia by calculating the integrated band intensities of NH₃ at 3 and 10 K before and after the reaction take place, respectively. The ratio between the two calculated band intensities allows us to deduce the amount of reacting ammonia. Table 2 gathers the integrated band intensities at 3 and 10 K of the ν₂ band of ammonia in a reactive sample containing N, N₂ and NH₃ species and in a reference sample containing only N₂ and NH₃ molecules as reactants.

Table 2 Integrated band intensities of ammonia at 3 and 10 K for NH₃ + N + N₂ and NH₃ + N₂ reactions S (cm⁻¹). The ratio between the integrated band intensities allows us to easily estimate the NH₃ amount consumed during the NH₃ + N reaction

	S 3K (cm^−1)	S 10K (cm^−1)	S 10K/S3K
NH3 + N + N2	0.85	0.76	0.89
NH3 + N2	0.79	0.81	1

---

The calculations of the integrated band intensities of ammonia at 3 and 10 K in the reference sample show almost the same value around 0.8 cm⁻¹, which confirms the fact that there is no NH₃ consumption and thus no reaction between molecular nitrogen and ammonia. We note that in the reactive sample containing nitrogen atoms the integrated band intensities of NH₃ are 0.85 cm⁻¹ and 0.76 cm⁻¹ at 3 and 10 K, respectively. This shows that NH₃ concentration is decreasing at 10 K; thus a reaction is occurring between NH₃ and nitrogen atoms and the NH₂ radical formation is directly linked to the presence of N-atoms in our solid samples. The ratio between the integrated band intensities at 3 and 10 K easily allows the estimation of the NH₃ amount consumed during the N + NH₃ reaction, which is found to be 0.89, showing that almost 11% of NH₃ has reacted with nitrogen atoms to produce NH₂.

IV. Discussion

We have characterized the NH₃ + N reaction in the solid phase, at very low temperature, under non-energetic conditions and where the NH₃ and N reactants are in the ground state. We have shown, by carrying out our experiments at temperatures as low as 3 K, that the first step of the NH₃ + N reaction may be blocked due to the reduced mobility of the trapped species in the solid phase. Leaving the formed solid sample in the dark for a while, to allow all trapped reactants to relax down to the ground state, is the only possible way to prove that the studied N + NH₃ reaction starts from reactants in their ground states and without providing any additional energy to the system. The appearance of the NH₂ radical as a reaction product by inducing the mobility of N(⁴S)-atoms between 3 and 10 K translates to the fact that a hydrogen abstraction reaction from ammonia under non-energetic conditions may occur at very low temperature.

From a theoretical point of view, we did not find any calculation in the literature to characterize reactions involving ground or excited state nitrogen atoms with ammonia, while many theoretical investigations have been carried out to study similar reactions such as N + H₂O, N + CH₄ and N + CH₃OH.²² Based on those previous theoretical models describing reactions with N-atoms, all reactions involving excited nitrogen atoms N(²D/⁴P) are highly exothermic, with almost no activation energy, and reactions involving ground state nitrogen atoms N(⁴S) are found to be endothermic with very important activation energies. We suppose that NH₃ + N(⁴S) reaction is not an exception and that it may need a high additional energy supply to occur. However, in the present experimental study, we show that the N + NH₃ reaction may occur only by inducing a mobility of ground state nitrogen atoms at very low temperature, between 3 and 10 K. In fact, the heating of the sample from 3 to 10 K stimulates the mobility of ground state nitrogen atoms which may recombine through reaction 1 to the form metastable molecular nitrogen N₂(A).²³ This metastable nitrogen species, which is highly reactive,²⁴ can be a source of high energy, 6.2 eV.

N(⁴S) + N(⁴S) → N₂(A) (1)

N₂(A) → N₂(X) + 6.2 eV (2)

By forming metastable molecular nitrogen N₂(A), through N(⁴S)–N(⁴S) recombination, a series of reaction processes start occurring. In the gaseous phase and without any interaction with the environment, N₂(A) may relax down to the ground state (reaction (2)). However, in the solid phase, N₂(A) excited molecular nitrogen interacts with all the species present in its environment. Regarding, for example, the composition of one of our standard solid samples [NH₃ + N/N₂] formed at 3 K, we note that it is made up on average of a very high amount of ground state molecular nitrogen N₂(100%) and a few ground state nitrogen atoms N (8%) and NH₃ molecules (0.2%). In such a sample, the formed metastable molecular nitrogen N₂(A) may essentially react with ground state molecular nitrogen in the solid sample through two reaction pathways:

N₂(A) + N₂(X) → N₂(X) + N₂(A) (3)

N₂(A) + N₂(X) → N₂(X,v′) + N₂(X,v) (4)

Metastable N₂(A) may also react with ground state nitrogen atoms to form excited nitrogen atoms.^25,26 Indeed, many studies characterizing discharge flows in nitrogen gas have shown that there is a favorable energy transfer process between metastable molecular nitrogen N₂(A) and ground state nitrogen atoms (reaction (5)), leading to the formation of highly excited²⁵ nitrogen atoms N(⁴P) and also to the first excited²⁶ atomic N(²D).

N₂(A) + N(⁴S) → N(⁴P/²D) + N₂(X) (5)

Finally all these nitrogen excited species derived from the mobility of ground state nitrogen atoms, namely N₂(A), N(⁴P) and N(²D), may react under non-energetic conditions with NH₃ to create reactive radical species (reactions (6) and (7)) such as NH and NH₂.

N(⁴P/²D) + NH₃ → NH + NH₂ (6)

N₂(A) + NH₃ → N₂(X) + NH₂ + H (7)

As mentioned previously, all these reaction pathways leading to the formation of NH and NH₂ radicals have still not been theoretically investigated and in particular they need computational explorations. These experimental results show that the NH₃ dissociation does not take place exclusively through energetic irradiations; it may also involve other reaction pathways such as NH₃ + N. This conclusion may be consistent with that of Takano studies³ concerning the variations of NH₃ abundances in some nearby galaxies. They showed that the ammonia low abundances might not only be due to photo-dissociation effects and that other chemical processes might be involved. In fact, various icy bodies in space are the site of radical-molecule reactions in the solid phase at very low temperature under non-energetic conditions, circumstances comparable to those shown in the present study. In this context, as for the reactions involving high particle energies, the chemistry involving N-atoms may also be a source of the increasing chemical complexity in various cold regions of the Universe, since many investigations have already shown that large amounts of atomic nitrogen^16,17 may exist in the core of some dark clouds and in such a situation N(⁴S)–N(⁴S) recombination processes leading to the formation of highly excited nitrogen species to initiate the destruction of NH₃ molecules might be very frequent. Such reaction pathways are of paramount importance for the formation of very reactive species such as NH and NH₂ radicals, which has an effect on the life cycles and abundances of ammonia. However, in the context of the present study we have investigated the N(⁴S) + NH₃ reaction in a specific environment which is a solid nitrogen and in a limited temperature range in order to control the reactant mobility to favor the formation of the NH₂ radical. As we are interested in solid-phase reactions relevant both to the interstellar medium and planetary atmospheres, we plan to investigate in the future the influences of water molecules and the temperature on the N(⁴S) + NH₃ reaction. Many interstellar icy grains are water-rich ices characterized by different temperatures and this may lead to different reaction pathways. Interaction between N-atoms and NH₃–H₂O mixtures may be a source of other products in addition to the NH₂ radical and NH₃ + N/N₂ co-depositions in the 70–200 K temperature range allow us to study reactions without interaction with solid N₂.

V. Conclusions

We show through this experimental study that, in the solid phase and at cryogenic temperatures, the NH₃ + N(⁴S) reaction might be one of the sources of the amidogen radical NH₂ in some cold regions of the interstellar space, in addition to those usually discussed which involve high energy photons and particles. Based on our obtained results, we propose reaction pathways to explain the NH and NH₂ radical formation which may be the first step in the activation of stable species such as NH₃, a significant process in the molecular complexity of the Universe. For the present study, the reactants have been frozen at a very low temperature and therefore their interactions have been controlled by stimulating the mobility of nitrogen atoms through a gradual heating of the solid sample in order to show that a hydrogen abstraction reaction from NH₃ may occur without supplying any external energy to the NH₃ + N system.

References

1. M. D. Hofstadter and D. O. Muhleman, Icarus, 1989, 81, 396.
2. M. J. Loeffler, U. Raut and R. A. Baragiol, Astrophys. J., Lett., 2006, 649, 133.
3. S. Takano, T. Takano, N. Nakai, K. Kawaguchi and P. Schilke, Astron. Astrophys., 2013, 552, 34.
4. J. Keene, G. A. Blake and T. G. Phillips, Astrophys. J., Lett., 1983, 271, 27.
5. A. C. Cheung, D. M. Rank, C. H. Townes, D. D. Thornton and W. J. Welch, Phys. Rev. Lett., 1968, 21(25), 1701.
6. T. J. Millar, A. Bennett, J. M. C. Rawlings, P. D. Brown and S. B. Charnley, Astron. Astrophys., Suppl. Ser., 1991, 87, 585.
7. E. F. van Dishoeck, D. J. Jansen, P. Schilke and T. G. Phillips, Astrophys. J., Lett., 1993, 416, L83.
8. P. Hily-Blant, S. Maret, A. Bacmann, S. Bottinelli, B. Parise, E. Caux, A. Faure, E. A. Bergin, G. A. Blake, A. Castets, C. Ceccarelli, J. Cernicharo, A. Coutens, N. Crimier, K. Demyk, C. Dominik, M. Gerin, P. Hennebelle, T. Henning, C. Kahane, A. Klotz, G. Melnick, L. Pagani, P. Schilke, C. Vastel, V. Wakelam, A. Walters, A. Baudry, T. Bell, M. Benedettini, A. Boogert, S. Cabrit, P. Caselli, C. Codella, C. Comito, P. Encrenaz, E. Falgarone, A. Fuente, P. F. Goldsmith, F. Helmich, E. Herbst, T. Jacq, M. Kama, W. Langer, B. Lefloch, D. Lis, S. Lord, A. Lorenzani, D. Neufeld, B. Nisini, S. Pacheco, T. Phillips, M. Salez, P. Saraceno, K. Schuster, X. Tielens, F. van der Tak, M. H. D. van der Wiel, S. Viti, F. Wyrowski and H. Yorke, Astron. Astrophys., 2010, 521, 52.
9. P. A. Gerakines, W. A. Schutte and P. Ehrenfreund, Astron. Astrophys., 1996, 312, 289.
10. W. Zheng, D. Jewitt, Y. Osamura and R. I. Kaiser, Astrophys. J., Lett., 2008, 674, 1242.
11. D. E. Miligan and M. E. Jacox, J. Chem. Phys., 1965, 43, 4487.
12. O. Schnepp and K. Dressler, J. Chem. Phys., 1960, 32, 1682.
13. N. Nishi, H. Shinohara and T. Okuyama, J. Chem. Phys., 1984, 80, 3898.
14. M. J. Loeffler and R. A. Baragiola, J. Chem. Phys., 2010, 133, 214506.
15. E. L. Zins and L. Krim, RSC Adv., 2013, 3, 10285.
16. A. Waszczak, 2013. Hydrogen and nitrogen cosmochemistry: a review. Ge 232 term paper.
17. D. C. Knauth, B.-G. Andersson, S. R. McCandliss and H. W. Moos, Nature, 2004, 429–636.
18. E. L. Zins, P. R. Joshi and L. Krim, Mon. Not. R. Astron. Soc., 2011, 415, 3107.
19. C. Pirim, L. Krim, C. Laffon, P. Parent, F. Pauzat, J. Pilmé and Y. Ellinger, J. Phys. Chem. A, 2010, 114, 3320.
20. C. R. Richey and P. A. Gerakines, Astrophys. J., 2012, 759, 74.
21. C. J. Bennett, C. Jamieson, A. M. Mebel and R. I. Kaiser, Phys. Chem. Chem. Phys., 2004, 6, 735.
22. C.-M. Ouk, N. Zvereva-Loete, Y. Scribano and B. Bussery-Honvault, J. Comput. Chem., 2012, 33, 2211.
23. A. M. Pravilov, L. G. Smirnova and A. F. Vilesov, Chem. Phys. Lett., 1984, 109, 4.
24. V. Guerra, P. A. Sa and J. Loureiro, J. Phys. D: Appl. Phys., 2001, 34, 1745.
25. D. Kuszner and N. Schwentner, J. Chem. Phys., 1993, 98, 9.
26. H. Kajihara, F. Okada and S. Koda, Chem. Phys., 1994, 186, 395.

---