Effect of electron temperature and concentration on production of hydroxyl radical and nitric oxide in atmospheric pressure low-temperature helium plasma jet: Swarm analysis and global model investigation

2.1 Model description

To obtain the better insight into reaction mechanisms for the production and consumption of OH and NO, we use zero-dimensional global model described in detail in our previous work [17], based on 1,488 reactions and 74 species for the mixture He/humid air. The term “zero-dimensional” model refers to the fact that it calculates chemical composition and analyzes the chemical production and consumption pathways of each plasma species in one fixed point of the discharge. A model is based on numerical solving of the system of coupled particle balance, electron concentration balance, and electron temperature balance equations as follows [12]:

where t, n i , n j … n f denote time and number density of ith, jth … and fth species, respectively, including electrons. k jm ( S ) and k ir ( L ) denote rate coefficients for two-body reactions between jth and mth species and between ith and rth species. Analogously, k lpq ( S ) and k isf ( L ) denote rate coefficients for three-body reactions between lth, pth, qth species, and ith, sth, fth species. Superscripts “ ( S ) ” and “ ( L ) ” stand for ith species “source” and “loss” process, respectively. F i represents the flow of i th species and superscript “ fgc ” stands for the “feed gas component,” for which an additional source term is included in Eq. (1), which describes the flow from the inlet of the system at starting time of the simulation ( t = 0 ) . The flow term is not included in Eq. (1), only in the case of electrons. S abs represents the absorbed power from electric field per unit volume of the plasma system (power density). The second and the third terms describe electron energy losses through elastic momentum transfer scattering on heavy particles and through various inelastic scattering processes. k eg el and k ej inel denote rate coefficients for electron momentum transfer scattering on i th plasma component and specific inelastic scattering on j th component, with concentrations n gi and n j , respectively. ε thr ej denotes energy loss (threshold) for a specific inelastic electron scattering process on the j th plasma component. In the case of superelastic collisions, the third term on the right side of Eq. (2) is used with the “+” sign.

Other calculation methods characteristic for fluid or hybrid models include the analysis of the spatial inhomogeneity of the plasma by solving the system of the fluid equations, and they are related with much more of computational burden. Solving the system of coupled particle balance equations in one fixed point of the plasma allows the insight into chemistry of the very complex chemical composition, which is characteristic for plasmas formed by mixing of He/Ar discharges with open air, with much less computational burden and time. To describe the changes in plasma characteristics and chemical composition along the symmetry axis of the plasma jet, one should use a “plug-flow” approach based on models for the power density, gas temperature, flow velocity, and air fraction obtained by fitting the experimental results, as presented by Van Gaens and Bogaerts [12]. The main goal of this work was not to analyze how the plasma properties change along the axis of the jet but to calculate the concentration of the OH radical and NO at fixed positions and to analyze the influence of electron temperature and concentration inside plasma bullet on production mechanisms for these important precursors for other reactive species.

With the intention of using the various measured values of electron temperature and concentration from the literature [14,15,16], in this work, we use the global model in the reduced form, and only the equations describing the balance of particle number density for each plasma species are included in calculation. The system of coupled particle balance equations is now expressed as follows:

where t and n i denote time and number density of ith species, respectively. S i , R and L i , R denote the total rates of source ( S ) and loss ( L ) of ith species through specific volumetric chemical reactions ( R ) , including two-body and three-body processes. F i represents the volume ( V ) averaged flow-in and flow-out term of i th species as in the study by Mladenović and Gocić [17] in units [ cm 3 / s ]. Outlet flow term is included in particle balance equations for all species, while the inlet flow term at starting time of the simulation is included only for feed gas components ( fgc ) . Model includes only gas phase processes due to the lack of data for particle-surface reaction probabilities in atmospheric pressure plasmas, and most of them are measured in low-pressure systems. They are also dependent on experimental conditions such as gas temperature and pressure, mixture, wall material, and surface conditions [53]. In this work, we analyze the production of OH and NO only in the gas phase inside the plasma channel of the jet in which plasma bullets propagate.

We analyze the chemical kinetics of OH and NO for the case of helium jet which propagates in open air, without any contact with metallic or dielectric surfaces. Also, as stated by Schröter et al. [53], an important fact is that probabilities of particle-surface reactions for atmospheric pressure plasmas are rare, most of them measured in low-pressure systems. They are also dependent on experimental conditions such as gas pressure, temperature or mixture, wall material, and surface conditions. Since we analyze the chemical kinetics of OH and NO only in gas phase, and due to the lack of data for probabilities and rates of particle-surface processes for atmospheric pressure plasmas, our global model excludes surface reactions. One should bear in mind that they could be important when modeling some plasma system in contact with metallic or dielectric surfaces.

During simulation, we calculate the percent of contribution for each process in generation and loss for all specific species by dividing the rate of the process S i , R and L i , R with total, summarized rate ∑ R S i , R and ∑ R L i , R . According to these statistics, we have extracted the main processes for the production of OH and NO at different conditions depending on electron temperature and electron density at different air and water vapor contents. The values of these parameters are estimated from the literature [14–16,24,25,39,46].

As presented in the studies by Mladenović and Gocić [17], we use BOLSIG+ for the calculation of EEDFs and rate coefficients for electron impact processes separately from the global model. As an effort to realize the coupling between BOLSIG+ and MATLAB calculation, in the first step, we calculate rate coefficients with the mole fractions estimated from the initial gas mixture, for each case of varied air and water vapor contents and run a global model. In the second step, we use the new mole fractions in BOLSIG+ obtained by a global model after 15 ms when our calculation reaches the steady-state conditions and the concentration of the most abundant plasma species, helium, nitrogen, oxygen, and water vapor, with the most pronounced influence on EEDF, remain constant during calculation, same as initial, for each case of varied air and water vapor contents. In numerical solving of BE in BOLSIG+ input cross-sectional data for electron – neutral scattering processes are multiplied by mole fractions for each of the included species [47]. In the next step, we use the new calculated rate coefficients and run the global model to check plasma composition and mole fractions of included species. We have done a few such iterations until reaching the good agreement, meaning that differences in mole fractions obtained by global model and used in BOLSIG+ calculations are at second decimal places or less.

The parameter for calculation of EEDF in BOLSIG+ is the reduced electric field E/N or the corresponding mean electron energy (represented by the electron temperature, although cold atmospheric pressure plasma is not in the equilibrium state). The rate coefficients for electron-neutral scattering processes can be represented as a function of E/N or mean electron energy. We use the second case because one of the important parameters that we vary in this work is electron temperature.

In some cases [14], rate coefficients for electron impact processes are taken in the Arrhenius form with the assumption that electrons have a Maxwell–Boltzmann (MB) distribution at the corresponding temperature. According to our recently presented results [42], when running the global model with rate coefficients obtained from MB EEDF differences of a few orders in magnitude can emerge for concentration of plasma species in comparison to those calculated with rate coefficients obtained from BE EEDF. Differences in EEDF, the most importantly in the high energy tail, strongly affect the rate of high threshold processes such as excitations, dissociations, and ionizations.

The system of Eq. (3) is solved by MATLAB ode15s solver, with relative and absolute tolerances equal to 10⁻¹⁰ and 10⁻⁶, respectively. A similar calculation procedure was used in our previous works [17,44], In our model, calculations are made with time step of 50 ns up to 15 ms as the final time during which our calculation reaches the steady state. The gas temperature during calculation was 296 K and the flow rate was 5 slm.

3.1 Effect of electron temperature on production of OH and NO

Figure 1a and b present BE EEDF calculated for the constant electron concentration 10¹⁰ cm⁻³ and two values for T e 1 and 4 eV, respectively. Different amounts of air and water vapor in helium plasma dramatically change the shape of EEDF at 1 eV and its high-energy tail. For 100 ppm of air in mixture, the population of high energy part of EEDF strongly increases with the increase of H₂O amount from 100 to 10,000 ppm. Presented results show that a higher amount of air rises the energy tail of EEDF for the same mean electron energy, requiring higher electric field due to more pronounced electron energy loss in scattering processes. Water vapor has an analogous effect on EEDF, which is more pronounced at lower mean energies as presented in Figure 1a for 1 eV than for 4 eV in Figure 1b.

Figure 1

EEDF calculated by BOLSIG+ for (a) mean electron energy 1 eV and (b) 4 eV, for a different amount of air and water vapor in helium plasma and n e = 10¹⁰ cm⁻³.

As a consequence, rate coefficients for the electron impact processes are increased at higher content of air and water vapor, and the effect is more pronounced at lower mean electron energies, as presented in Figure 2 for dissociative electron attachment to H₂O (a), dissociative attachment to O₂ (b), electron impact dissociation of H₂O (c), O₂ (d), N₂ (e), and ionization of O₂ (f). These processes are important for chemical kinetics of reactive species since dissociative attachment to H₂O and O₂ can be regarded as a first step of reaction chain for producing OH (Section 3.1), electron impact dissociation of O₂ and N₂ initially produce O and N as a precursor for the production of OH and NO, respectively, while the ionization processes are important for production of hydrate-cluster ions, which also play a significant role in OH production after few milliseconds when plasma chemistry becomes very complex. In the previous work [17], we have shown that water vapor profoundly increases plasma electronegativity, while in this work, we show that dissociative attachment to H₂O molecules can be also very important for the production of OH radicals at lower mean electron energies.

Figure 2

Rate coefficients calculated by BOLSIG+ for (a) electron attachment to H₂O, (b) electron attachment to O₂, (c) dissociation of H₂O, (d) dissociation of O₂, (e) dissociation of N₂ and ionization od O₂, and (f) for a different amount of air and water vapor in helium plasma and n e = 10¹⁰ cm⁻³.

Figure 3 presents the concentrations of OH calculated by global model at constant electron temperatures 1, 2, 3, and 4 eV and at different amounts of air and water vapor. The electron concentration is kept constant at 10¹⁰ cm⁻³ during calculation to analyze the influence of electron temperature on the production of OH. The optimal values of air concentration are chosen from the results of LIF measurement presented in the literature [24] depending on the axial position in the plasma jet. Also, two different concentrations of water vapor are chosen as a representative for the cases when H₂O comes only from the humidity of ambient air (100 ppm of H₂O), or it is included as feed gas component for reinforced production of OH (5,000 ppm of H₂O).

Figure 3

Concentration of OH in plasma as a function of electron temperature for different amounts of air and water vapor in helium plasma and n e = 10¹⁰ cm⁻³ after 15 ms.

As presented in Figure 3, for a low amount of air in plasma (dashed lines with scatter), the concentration of OH radical is increased with mean electron energy despite the amount of water vapor. At low energies, OH concentrations for 100 and 5,000 ppm of H₂O differ more than for high energies since the rate coefficient for electron impact dissociation of H₂O is not influenced by water vapor above 3 eV (Figure 2b). For higher amount of air (solid lines with scatter), chemical kinetics of OH radical is more complex, and there are more important production processes besides electron impact dissociation of H₂O, also depending on mean energy, as presented in Figure 4.

Figure 4

The main production processes of OH radical as a function of mean electron energy for 10,000 ppm of air with 100 ppm H₂O (a) and 5,000 ppm of H₂O (b). Electron concentration is kept constant n e = 10¹⁰ cm⁻³ during calculation for 15 ms.

The production of OH radical is initially determined by dissociative electron attachment at 1 eV. In the case of 100 ppm of water vapor, the attachment to O₂ (Figure 4a) is dominant over attachment to H₂O through process O⁻ + H₂O → OH + OH⁻ and the pathway that includes H 2 O 2 − ions produced by the process O⁻ + H₂O + He → H 2 O 2 − + He. After 15 ms, the composition of the plasma becomes more complex, and chemical reactions of O₂(¹ ∆ ) and O with HO₂, produced by process H + O₂ + He → HO₂ + He, become important in the production of OH. At higher mean electron energies and 100 ppm of H₂O, the important pathways are reactions of excited oxygen species O(¹D) and O(¹S) with H₂O molecules due to the increased total rate for electron impact dissociation of O₂. Ion conversion H₂O⁺ + H₂O → H₃O⁺ + OH and process involving hydrate H₃O⁺ · OH, produced in O 2 +  · H₂O + H₂O → H₃O⁺ · OH + O₂, are important to produce OH only at high energies (4 eV) when the total rate for electron impact ionization of O₂ and H₂O is high enough. The same is valid for the process O 2 +  · H₂O + H₂O → O₂ + OH + H₃O⁺, which is, beside NO₃ + H → NO₂ + OH and NO₃ + HO₂ → O₂ + NO₂ + OH, included in the pattern part of bars in Figure 4a). When the water vapor is included as a feed gas component (5,000 ppm), the chemical kinetics of OH radical is determined with the same processes (Figure 4b) but percents of contribution in the total OH production differ than at 100 ppm of H₂O for a several important reasons, apart from different plasma composition. As the first, rate coefficients for electron impact dissociation producing OH, O, O(¹D), and O(¹S) are influenced by H₂O concentration through modification of EEDF energy tail, as indicated in Figure 1. These effects are even more enhanced in the case of processes with a higher energy threshold such as the ionization of H₂O and O₂, producing ions and hydrates. Second, the influence of water vapor on EEDF and rate coefficients is more pronounced at lower mean electron energies (Figures 1 and 2). As a consequence, dissociative electron attachment to H₂O molecules has the dominant role in the production of OH at 1 eV with more than 60% of contribution and remains important even at higher energies when processes involving excited states O(¹D), O(¹S), and O₂(¹ ∆ ) produce OH. Electron impact dissociation of H₂O becomes important above 2 eV, and dominant over the dissociative attachment at 3 and 4 eV. In our model, processes that are included in the pattern part of bars at these energies involve negative ion hydrates: H₂O⁺ + OH⁻ · H₂O + He → H₂O + OH + H₂O + He and O 2 +  · H₂O + OH⁻ · H₂O + He → 2H₂O + He + O₂ + OH. This also indicates the importance of dissociative attachment to H₂O for the production of OH since these hydrates are mainly produced in the convolution of processes, first H₂O + e⁻ → H⁻ + OH, second H⁻ + H₂O → OH⁻ + H₂, and third OH⁻ + H₂O + He → OH⁻ · H₂O + He, with the total rate proportional to the third power of water vapor concentration in plasma. Finally, according to results presented in Figure 3, concentration of OH at 10,000 ppm of air almost reaches saturation above 3 eV in the interval 10¹³–10¹⁴ cm⁻³, depending on the water vapor mole fraction, due to the increased rate of electron impact dissociation of N₂ molecules and process that convert OH radical into nitric oxide OH + N → NO + H. Concentration of this reactive species is presented in Figure 5, for n e = 10¹⁰ cm⁻³ after 15 ms of calculation.

Figure 5

Concentration of NO in plasma as a function of mean electron energy for different amounts of air and water vapor in helium plasma and n e = 10¹⁰ cm⁻³.

In our model, nitric oxide is taken as a component of ambient air (10⁻⁶%) [17]. For the low amount of air effluent in plasma (dashed lines with scatter symbols), the concentration of NO profoundly rises with the mean electron energy, and it is more influenced by water vapor, as presented in Figure 5, than for 10,000 ppm of air (solid lines with scatter), when similar saturation occurs like in the OH case due to the conversion process OH + N → NO + H, which couples kinetics of OH and NO. For the purpose of biomedical applications, it is important to reinforce the production of NO, and the discharges with higher amounts of air are used [54]. Having this fact in mind, we have chosen to analyze the main production pathways for NO at 10,000 ppm of air, as presented in Figure 6.

Figure 6

The main production processes of NO as a function of the mean electron energy for 10,000 ppm of air with 100 ppm H₂O (a) and 5,000 ppm of H₂O (b). Electron concentration is kept constant n e = 10¹⁰ cm⁻³ during calculation for 15 ms.

At low amount of water vapor (Figure 6a) and low energies around 1 eV, electron impact dissociation of N₂ (9.7 eV) and conversion process OH + N → NO + H do not play an important role in the production of NO, which is then initially generated mainly through the effluent of air in plasma. After 15 ms of global model calculation, the composition of plasma becomes more complex, and we recognize new important generation pathways, which include oxygen atoms O + NO₂ → NO + O₂ and O + NO 3 − → NO + O 3 − . The first is a part of a loop of processes since NO generates NO₂ through two-body and three-body association with O, and the second implicates the importance of electron attachment at low energies since NO 3 − ions are mainly generated through charge transfer O⁻ + NO₃ → NO 3 − + O and three-body ion conversion O⁻ + NO₂ + He → NO 3 − + He. Beside chemical complicity, these processes become important for creation of NO also because the rate coefficient for dissociation of O₂ (4.5 eV) is a few orders of magnitude higher at 1 eV than for dissociation of N₂, as presented in Figure 2c–d. For mean electron energies of 2 eV or higher, dissociation of H₂O (7.6 eV) takes place in kinetics of NO through processes OH + N → NO + H and dissociation of O₂ through channel O + NO₂ → NO + O₂. At 4 eV, dissociation of N₂ is reinforced enough so that chemical reactions of excited species N(²D) and N(²P) with O₂ molecules also occur as important NO production channels.

If water vapor is included as a feed gas component, the most important pathway for the generation of NO is conversion OH + N → NO + H for 2 eV and higher energies, as presented in Figure 6b. Only at 1 eV, this process is competitive with the flow of NO from ambient air. Processes that are included in the pattern part of bars in Figure 6 are O₂ + N(²D) → O + NO, O₂ + N(²P) → O(¹D) + NO, O₂ + N(²P) → O(¹S) + NO, and N + NO₃ → NO + NO₂, with a few percent of contribution in generation of NO for each, and are excluded from Figure 6 for the purpose of simplicity of presentation.

3.2 Effect of electron concentration on production of OH and NO

In Figure 7, we present the calculated EEDFs for 1% of air in helium plasma and for a different amount of water vapor and electron concentration. These values are taken as optimal from experimental data in the literature [24], based on LIF measurements for free plasma jet in open air and measurement of electron concentration and temperature for plasma bullets, which propagates in the plasma channel of the jet [14–16,55]. Distributions are also compared with the Maxwell–Boltzmann distribution for the same mean electron energy 1 eV (Figures 7a) and 4 eV (Figure 7b). For electrons behind the ionization front, Maxwellianization effects of EEDF are observable only at electron concentration 10¹⁴ cm⁻³, while for 10¹⁰ cm⁻³ and even 10¹² cm⁻³, high energy tail of nonequilibrium EEDF differs from MB tail for a few orders of magnitude. In the case of higher mean electron energies (Figure 7b), water vapor has a negligible influence on EEDF, and the effect of Maxwellianization is not observable even at electron concentration 10¹⁴ cm⁻³, which is measured as the highest in cold atmospheric pressure pulsed discharges [16]. As a consequence, rate coefficients for electron impact dissociation of O₂, N₂, and H₂O and other inelastic processes are increased by electron concentration at lower mean electron energies, but not influenced above 3 eV neither by electron concentration nor by water vapor. Similar results are presented in our previous work [42] for a micro-jet with a much lower and constant concentration of air and H₂O. In this work, we present the results of a more complex study for a wide range of discharge conditions based on experimental data presented in the literature.

Figure 7

EEDF calculated by BOLSIG + for 1% of air and (a) mean electron energy 1 eV and (b) 4 eV, with 100 ppm (blue lines) and 1,000 ppm H₂O (red lines) and for electron concentration 10¹⁰ cm⁻³ (solid lines), 10¹² cm⁻³ (dashed lines), and 10¹⁴ cm⁻³ (dotted lines). Solid black lines represent Maxwell–Boltzmann EEDF. (b) Calculated EEDFs overlap for a wide range of varied amounts of H₂O and n _e.

As it is presented in our previous work [42], the results of a system of coupled particle balance equations in 0D global model calculation are mostly dependent on the variation of rate coefficients with the mean electron energy (electron temperature), reflecting the influence of EEDF. Since the electron temperature has the specific space distributions in plasma bullets and depends on specific plasma discharge, according to the results presented in Figure 7, it can be expected that Maxwellianization effects on EEDF can introduce important differences in the production of OH radical and nitric oxide NO.

With rate coefficients obtained by BOLSIG+ for each set of conditions (water vapor and electron concentration), we have calculated plasma composition after 15 ms and made the analysis of all important chemical pathways for producing OH radical and NO. In Figures 8 and 9, we present the concentration of these reactive species for three different electron concentrations, 10¹⁰, 10¹², and 10¹⁴ cm⁻³, and two different amounts of water vapor, 100 ppm and 1,000 ppm, which represent the order of magnitude when water vapor comes only from humidity of ambient air or it is included as feed gas component. Figures also present estimated interval for measured concentrations of OH and NO from the literature [25,26,27].

Figure 8

Calculated OH concentration for 1% air in plasma and different concentrations of electrons and water vapor amounts. The estimated interval of measured OH concentration from the literature [24] is also marked.

Figure 9

Calculated NO concentration for 1% of air in plasma and different concentrations of electrons and water vapor amounts. The estimated interval of measured NO concentration from the literature [25,26,27] is also marked.

To verify our model, we have made the comparison of our calculated results for OH and NO concentration with experimental data from the literature, for the range of electron temperature 1–4 eV and electron concentration 10¹⁰–10¹⁴ cm⁻³ in Figures 8 and 9. Model reaches the good agreement with measured values of OH concentration with those presented by Yonemory et al. [24] at all electron temperatures for electron concentration below 10¹² cm⁻³, which was also reported by Yonemory et al. [24]. Concerning NO, above 1 eV, we reach good agreement with the experimental data in previous studies [25,26,27] for electron concentration above 10¹² cm⁻³, characteristic of pulsed discharges.

As presented in Figure 8, the calculated OH concentration rises with the electron temperature for all values of electron concentrations. At 1 eV, an increase in electron concentration from 10¹⁰ to 10¹⁴ cm⁻³ leads to an increase in OH concentration from around 5 × 10¹¹ to 3 × 10¹⁴ cm⁻³ (for 100 ppm H₂O) or 1 × 10¹² to 5 × 10¹⁴ cm⁻³ (for 1,000 ppm H₂O). At high energies (4 eV), an increase in electron concentration increases OH concentration from around 3 × 10¹³ to 3 × 10¹⁵ cm⁻³ (for 100 ppm H₂O) or 5 × 10¹³ to near 1 × 10¹⁶ cm⁻³ (for 1,000 ppm H₂O). These effects of n e and T e resemble the Maxwellianization effects on EEDF through the values of rate coefficients and thus total rates of OH production pathways as presented in Figure 4.

Having in mind that for biomedical applications the temperature of active medium should be near to body temperature, plasmas with lower mean electron energy are preferable. According to Eq. (2), it is hard to obtain the low electron temperature at high E / N obtained by high absorbed power. According to the results presented in Figure 8, the best agreement between our model results and experimental data is obtained for lower electron temperatures 1–2 eV and electron concentration of 10¹⁰–10¹² cm⁻³, also reported in the literature as a characteristic for many types of CAPS. So, one can conclude that dissociative attachment to H₂O molecules is very important to produce OH radicals in plasmas at lower mean electron energies.

In both cases, electron concentration and water vapor increase the concentrations of OH and NO, but the best agreement between global model results and measured data is obtained for electron concentration 10¹⁰ and temperature 2 eV for OH (Figure 8), and some higher values 10¹² cm⁻³ and 3 eV for NO (Figure 9), due to higher threshold for electron impact dissociation of N₂. These values are also in good agreement with measured electron concentration and temperature for free jet, which propagates in open air from the literature [14,15,16]. We chose to make an analysis of all important chemical pathways for the production of OH and NO at given conditions.

In the case of OH important production channels for 100 ppm H₂O, n _e = 10¹⁰cm⁻³ and T _e = 2 eV are presented in Figure 4a), while at 1,000 ppm H₂O, the most important processes are O(¹D) + H₂O → OH + OH (15.3%), O(¹S) + H₂O → OH + OH (22.2%), O + HO₂ → O₂ + OH (13.3%), and O₂(¹D) + HO₂ → O + O₂ + OH (32.3%). Dissociative electron attachment and dissociation of H₂O molecules contribute to OH production summary with only 6%.

In the case of NO, at 3 eV and electron concentration 10¹² cm⁻³, the most important production channels are O + NO₂ → O₂ + NO (22%/11.7%), N + OH → NO + H (22.2%/29.4%), and NO₂ + H → NO + OH (33%/49%) (percents of contribution are related to 100 and 1,000 ppm of H₂O, respectively).

At the same mean energy but higher electron concentration 10¹⁴ cm⁻³, concentration of NO reaches the order of magnitude 10¹⁵ cm⁻³ (Figure 9), and these processes contribute to total NO production with O + NO₂ → O₂ + NO (18%/14%), N + OH → NO + H (7%/7.4%), and NO₂ + H → NO + OH (46%/48%) at 100 ppm and 1,000 ppm of H₂O, respectively. Moreover, new processes including excited species are recognized as important for the production of NO: O(¹D) + NO₂ → O₂ + NO (11.9%/7.7%), and O(¹D) + N₂O → NO + NO (3.5%/8.5%), since the main O(¹D) production channel is the electron impact dissociation e⁻ + O₂ → O + O(¹D) + e⁻ with the total rate profoundly increased with electron concentration and also due to Maxwellianization of EEDF. One should have in mind that these processes occur as important after 15 ms of calculation and are the part of a complex chemical loop, while the initial production of NO is mainly carried by OH conversion with N and flow in from ambient air.

We have made the analysis of all important production and consumption processes for reactive species NO _x and HNO _x for a wide range of discharge conditions, which are presented schematically in Figure 10. The percent of contribution for each process depends on the chemical composition of plasma/humid air mixture and also electron temperature and concentration. As presented, nitric oxide initially comes from flow of air in the plasma and conversion of OH with nitrogen atoms, produced by electron impact dissociation of N₂, and then becomes an important precursor for other reactive nitrogen-oxygen compounds. After a few milliseconds of calculation, chemical composition of the plasma becomes more complex and new chemical pathways, and NO _x becomes important for not only the production of NO but also for the enhanced production of acids HNO _x through three-body association with H and OH, which are then carried out from the system through flow. An analogous scheme representing the chemical kinetics of OH radical with all important pathways is presented in in our paper [17].

Figure 10

The main production and consumption mechanisms for species NO _x and HNO _x , determined by a 0D global model after 15 ms of calculation.