Europe PMC Funders Group Author Manuscript Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21. Published in final edited form as: Plasma Sources Sci Technol. 2015 February 1; 24(1): . doi:10.1088/0963-0252/24/1/015029.
Europe PMC Funders Author Manuscripts
Chemistry in glow discharges of H2 / O2 mixtures. Diagnostics and modelling M Jiménez-Redondo*,1, E Carrasco1,2, V J Herrero1, and I Tanarro1 1Instituto
de Estructura de la Materia (IEM-CSIC), Serrano 123, 28006 Madrid, Spain.
Abstract The chemistry of low pressure H2 + O2 discharges with different mixture ratios has been studied in a hollow cathode DC reactor. Neutral and positive ion distributions have been measured by mass spectrometry, and Langmuir probes have been used to provide charge densities and electron temperatures. A simple zero order kinetic model including neutral species and positive and negative ions, which takes into account gas-phase and heterogeneous chemistry, has been used to reproduce the global composition of the plasmas over the whole range of mixtures experimentally studied, and allows for the identification of the main physicochemical mechanisms that may explain the experimental results. To our knowledge, no combined experimental and modelling studies of the heavy species kinetics of low pressure H2 + O2 plasmas including ions has been reported before.
Europe PMC Funders Author Manuscripts
As expected, apart from the precursors, H2O is detected in considerable amounts. The model also predicts appreciable concentrations of H and O atoms and the OH radical. The relevance of the metastable species O(1D) and O2(a1Δg) is analysed. Concerning the charged species, positive ion distributions are dominated by H3O+ for a wide range of intermediate mixtures, while H3+ and O2+ are the major ions for the higher and lower H2/O2 ratios, respectively. The mixed ions OH+, H2O+ and HO2+ are also observed in small amounts. Negative ions are shown to have a limited relevance in the global chemistry; their main contribution is the reduction of the electron density available for electron impact processes.
1. Introduction Low pressure plasmas in electrical discharges with H2 and O2 are of interest in a variety of fields. In astrochemistry, the formation of H2O and H3O+ is of great relevance as they can be detected in interstellar environments [1-3]. Ions containing oxygen and hydrogen are formed in interstellar clouds [4], and they are assumed to play an important role in gas phase chemical routes leading to the production of H2O. The hydronium ion (H3O+) has been observed in molecular clouds since the nineties [5]. The recent Herschel mission has led to the detection of OH+ [6] and H2O+ [7] in the diffuse interstellar medium and their role as markers of regions with a small fraction of H2 has been highlighted [8]. The HO2+ ion remains unobserved to date. It has been repeatedly considered as a possible tracer for
*
[emailprotected]. 2Present address: Lehrstuhl für Physikalische Chemie II, Universität Erlangen-Nürnberg, Egerlandstr.3, D-91058, Erlangen, Germany.
Jiménez-Redondo et al.
Page 2
Europe PMC Funders Author Manuscripts
molecular oxygen [9-11] in a similar way as N2H+ is for N2, but the available thermodynamic and kinetic data suggest that the concentration of this ion in the ISM should be too small to be detectable [10]. In fusion research, discharge cleaning is used to eliminate the residual molecules in a vacuum vessel, of which oxygen and water are major components [12, 13], and oxygen-containing cold plasmas have been proposed for the removal of co-deposits at the reactor walls [14, 15]. Hydrogen and oxygen plasmas are also widely used in surface treatments, like chemical modification [16-18], decontamination [19] or functionalization of carbon nanotubes [20]. Previous studies on this kind of plasmas have been carried out under different conditions and with very different objectives. Atmospheric pressure discharges have been used in experimental and theoretical studies of H2 + O2 ignition [21], in the simulation of gas heating processes in H2 + O2 streamers [22], or in the modelling of production mechanisms of different neutral species [23, 24]. An extensive global model for mixtures of He with small fractions (< 1%) of H2O has been elaborated by Liu et al. [25], and the formation of OH radicals in plasma assisted combustion of H2 / air mixtures has been studied experimentally and theoretically by Yin et al. [26]. Low pressure plasmas of H2 + O2, mostly in the mbar range, have been employed by various groups. They have been used in the infrared spectroscopy analysis of the spectrum of the H3O+ ion [27]. Nevertheless, kinetic studies of these plasmas are limited to the analysis of the neutral species, such as the determination of oxygen atom concentrations in microwave post discharges of He-H2-O2 mixtures using NO tritration [28], the modelling of neutral species in the afterglow of a H2 + O2 discharge [29], or the experimental study of the variation of the O2(a1Δg) concentration with the introduction of small amounts of H2 in an Ar + O2 microwave discharge [30].
Europe PMC Funders Author Manuscripts
In this work, we present a study of the chemistry of neutral and ionic species in H2/O2 plasmas based on the experimental diagnostics and kinetic modelling of hollow cathode discharges at a pressure of 8 Pa and mixture proportions ranging from pure H2 to pure O2. For the pressure value selected, which lies toward the high pressure limit of the stable operating range of the discharge (1-10 Pa), the ion distributions in the plasma are largely determined by ion-molecule chemistry, which is the goal of the present study. For the lowest operating pressures, displaying higher electron temperatures, the ion distributions tend to be dominated by the products of direct electron impact ionization [31]. Previous experimental results for H2/O2 plasmas at similar conditions but restricted to low O2 concentrations (up to 15%) were given in [32]. Langmuir probes provide values for the electron temperatures and densities, and neutral and positive ion concentrations are determined by mass spectrometry. The main surface and gas processes are identified by comparison of experimental data and model predictions, and their relative relevance under the different discharge conditions is analysed.
2. Experimental setup The DC plasma reactor used for the experiments has been described in previous works [33, 34]. It consists of a stainless steel grounded hollow cathode reactor (34 cm length, 10 cm diameter) with a central anode. The vessel is pumped to a base pressure of 10−4 Pa by a 300 l/s turbomolecular pump backed by a dry pump. The reactor has additional ports for pressure
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 3
Europe PMC Funders Author Manuscripts
gauges, windows and coupling of the experimental techniques for the plasma diagnostics. Neutral species from the plasma are sampled with a quadrupole mass spectrometer with electron impact ionization (Balzers, Prisma QMS 200). A plasma process monitor (Balzers, PPM421), which includes an electrostatic focusing system, a cylindrical mirror energy analyzer and a quadrupole mass filter, is used to measure the mass and energy distributions of the positive ions hitting the cathode. Both of these instruments are installed in a differentially pumped vacuum chamber and sample neutrals or ions from the plasma through a diaphragm of ~ 100 μm diameter. The pressure during plasma measurements in this chamber is ~10−5 Pa. Electron temperatures and charge densities are measured with double Langmuir probes made in our laboratory [35]. The electron mean temperature and total charge density in the reactor are obtained from the analysis of the characteristic curves of the double Langmuir probe measurements in each discharge. The condition of orbital limited motion in a collision-free probe sheath is considered to be met [36]. The use of a single probe was also attempted in our reactor, but it was not possible to reach the electron saturation current due to the appearance of a secondary glow discharge in the probe when its positive potential was increased above the plasma potential.
Europe PMC Funders Author Manuscripts
Mixtures of H2/O2 of different proportions (from pure H2 to pure O2), with a total pressure of 8 Pa, are used for plasma generation. Measurements are performed under continuous flow conditions, and the pressures in the chamber (measured by a capacitance manometer Leybold CTR90) are regulated by balancing the gas flow with needle valves at the entrance (one for each gas) and a butterfly valve at the exit of the reactor, whose position is kept fixed during the experiments. For neutral species, calibration of the different sensitivity of detection of the QMS 200 for each pure gas is performed by comparing the capacitance manometer readings at various known pressures and the mass spectrometer ones. The H2O concentration is obtained by comparison with the calibration of the QMS for Ne (due to the similar mass), which is done in the same way as for O2 and H2, but taking into account the necessary corrections due to the different ionization cross section and the isotopic abundance of Ne. Calibration of the PPM421 for ions is performed with the noble gases He, Ne and Ar. The plasma monitor is used in the neutral detection mode, and the signal of each gas is compared to the corrected readings of a Bayard-Alpert gauge located in the same chamber. Taking into account possible errors in the determination of efficiencies for the various neutrals and ions and the reproducibility of the measurements, the uncertainties in the experimental concentrations of the different species are estimated to be ~ 20%. The residence time of gases in the reactor is measured according to the method used in previous works [33], obtaining a value of 0.54 ± 0.10 s. Plasma currents of ~ 150 mA and voltages of 500–550 V (depending on the mixture ratio) are maintained during the experiments. An electron gun is used for plasma ignition and then turned off.
3. Model A zero order kinetic model is employed for the interpretation of the experimental results. It is based on a set of coupled differential equations describing the time evolution of the
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 4
Europe PMC Funders Author Manuscripts
concentrations of both neutral and ionic species from the ignition of the discharge until the attainment of the steady state. Similar models applied to H2+D2, H2+N2 and H2+Ar discharges can be found in refs. [37-40]. The model accounts for the main physico-chemical processes: electron impact dissociation and ionization, ion-molecule and ion-ion reactions, neutralization at the wall and heterogeneous chemistry. The full set of considered reactions can be found in Tables 1 and 2. Electron impact reactions have rate coefficients which depend on the electron temperature. These rate coefficients have been calculated from the corresponding cross sections assuming a Maxwellian distribution for the electron energy distribution function, with a temperature that is derived from the Langmuir probe measurements. Other parameters required by the model are charge density, residence time and H2/O2 ratio. Unless otherwise indicated, experimental values for all of these parameters have been used as input for the model.
Europe PMC Funders Author Manuscripts
The assumption of a Maxwellian distribution is only an approximation. The actual shape of the electron energy distribution function (eedf) is not known with precision in our plasmas. On the one hand, inelastic and reactive collisions can deplete the high energy tail of a thermal eedf [75-77]. On the other hand, a small amount of non-thermal electrons, with energies extending to that corresponding to the cathode-anode voltage, can also be present in the discharge [78-80]. Electron impact dissociative processes can give rise to hot atoms that are mostly thermalized within the glow. In the steady state, a small fraction of these hot atoms survives and can be detected either in the line profiles of Balmer emission lines of H atoms [81] or indirectly in the energy resolved mass spectra of H+ and O+ ions [81-83]. However, previous works carried out on this reactor for different gas mixtures at comparable pressures [33, 34, 36-39] have shown that the assumption of a Maxwellian distribution is not a bad approximation for the global kinetics, and we have adopted it here for lack of better information on the actual eedf. The limitations of this approximation should be taken into account in the interpretation of the results. As in previous works, a gas temperature Tg = 300 K has been assumed in the model calculations for the cold plasmas in our room temperature reactor. Rotational temperature estimates, based on emission spectroscopy for comparable plasmas [81, 84], set an upper limit of 350-400 K for the gas temperature. Both neutral and charged species have been included in the model. Among neutrals, we can make a distinction between stable neutral species (H2, O2 and H2O) and radicals. Experimental data is only available for the first group, allowing for comparison with the model. Two metastable species, O(1D) and O2(a1Δg) (also referred to as O2(a)), have been included in the model due to the importance of their gas phase neutral reactions and their relevance in the formation of negative ions, respectively. Vibrationally excited states of H2 have not been contemplated explicitly in the model. In a previous study of pure H2 plasmas in the same reactor, emission spectroscopy measurements in conjunction with a collisional radiative model [33] showed that the H2 vibrational populations in our plasmas are concentrated in the lowest levels and can be roughly described by a vibrational temperature of ~ 3000 K. For this vibrational temperature, the
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 5
population of H2(v≥1) is ~ 12%. Given the high threshold for electron impact dissociation of H2 (~ 11 eV) as compared with the first vibrational quantum of H2 (~ 0.5 eV), we do not expect a significant contribution of vibrationally excited molecules, H2(v), to the global electron impact dissociation of H2. The decrease in population with growing v is far more important than the increase in the rate coefficient due to the lower energy threshold.
Europe PMC Funders Author Manuscripts
Charged species comprise positive and negative ions and electrons, which have distinct destruction mechanisms. Positive ions are assumed to disappear from the plasma mainly due to diffusion through the sheath to the reactor walls, where they are neutralized with a probability of 1 (see ref [34] for details on the treatment of ion diffusion). To meet the electroneutrality condition, the flux of positive ions to the walls is balanced to match the net production of positive charge in the glow [33]. Other mechanisms such as electron impact neutralization and ion-ion recombination are also considered, but their importance is orders of magnitude lower. For negative ions, collisional detachment with neutrals is the main destruction process, although electron impact detachment has also been included along with the above mentioned ion-ion recombination. Negative ions are considered to be confined within the discharge by the space charge electric field and, as assumed in other models [45, 85] do not diffuse to the walls. Total charge density is assumed to remain constant, so the value for the electron density is obtained by subtracting the total negative ion density from it.
Europe PMC Funders Author Manuscripts
Neutral chemistry is driven mainly by electron impact dissociation and wall recombination, although some gas phase bimolecular reactions (often including O(1D)) are also important. Heterogeneous processes at the wall surface are responsible for the formation of H2O, apart from the regeneration of the parent molecules H2 and O2. Table 2 shows the heterogeneous reactions included in the model, as well as the corresponding rate coefficients γ. The kinetic description for heterogeneous reactions follows the formulation employed in refs. [37, 38]. A detailed account of the treatment of diffusion of neutral species to the wall in the model can be found in ref. [34]. In agreement with these studies, the flux of atoms (O and H) and radicals (OH) to the surface wall dominate the heterogeneous chemistry including both adsorption and recombination via Eley-Rideal (ER) reactions. The γER selected for H2 is 3.5×10−3 (taken from reference [39]) and a value of 2.4×10−2 is estimated for O2, above the value considered for N2 in [39]. This assumption is based on results from [86], where higher wall recombination coefficients are obtained from O radicals than from N radicals on stainless steel. Reactions W8 to W10 are responsible for the formation of water at the reactor wall through ER reactions. Water production can require the formation of OH(s) firstly (W6W7), which is then removed by impinging H atoms (W8); or it can be generated by direct abstraction of H(s) or O(s) atoms with OH radicals and H2 molecules, respectively (W9 and W10). The related γER values (not available in the literature) have been chosen for a best global agreement with the data measured in this work over the whole range of studied concentrations. Dissociative adsorption of O2 at the walls, which has been experimentally studied by molecular beams and modelled on clean Fe surfaces [87, 88], has been tested with model simulations and has very small influence on the present H2 / O2 plasmas. In previous works [37-39], adsorptions of neutral molecules were considered negligible as compared with those of atoms and radicals. Literature calculations suggest that the
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 6
Europe PMC Funders Author Manuscripts
adsorption of water on surface iron atoms can take place without a barrier [89, 90]. However, model simulations with this assumption, including the subsequent reactions of adsorbed water, lead to relevant discrepancies with our experimental results, which are accounted for much better by assuming no H2O adsorption. Likewise the LangmuirHinshelwood heterogeneous recombination of adsorbed reactants like OH(s) + H(s) or 2 OH(s) for water production, with predicted energy barriers of ~ 1.1 eV and ~ 0.65 eV [89], respectively, have been considered minor, as compared with the barrierless Eley-Rideal processes listed in table 2 and have not been included in the final modelling.
4. Results and discussion The electron temperatures and densities measured with the Langmuir probe (along with the model values, which will be discussed below) can be found in figure 1. The values depend on the relative fraction of O2 in the precursor mixture (O2/(O2+H2)). Electron temperatures are stable through the range of mixtures investigated, with a mean value of ~ 2.4 eV, except for the pure hydrogen discharge, where it falls to ~ 1.7 eV. The special behaviour of the pure H2 discharge is also reproduced for the electron densities, where for pure hydrogen the value is higher than those for the rest of the mixtures. The higher electron temperatures measured for the mixtures containing O2 indicate that the high energy tail of the eedf is more populated. This results in a higher ionization efficiency and, as a consequence, less electrons would be needed to maintain the discharge, as observed. On the other hand, it seems that, once a small amount of O2 is present in the mixture, electron temperature and density are determined by the total gas pressure rather than by the H2/O2 proportion. The reasons for this behavior are not obvious and in any case lay beyond the scope of this work.
Europe PMC Funders Author Manuscripts
The experimental results for the relative concentrations of the neutral stable species species O2, H2 and H2O, normalized to the sum of concentrations of just these species (which are the only three that can be detected by our experimental set-up) are represented as symbols in figure 2. Water is produced in the discharge in appreciable amounts, reaching a maximum of ~ 30% of the total neutral concentration for a mixture with ~ 40% O2, close to what would be expected from a stoichiometric point of view, which would correspond to maximum water formation for a fraction of oxygen of 33%. The oxygen precursor is depleted at high H2 fractions, as it is mostly dissociated and then recombined at the wall to form H2O. The H2 precursor is also depleted at high oxygen fractions, but in a smaller proportion. Other neutral stable species, such as O3, were not detected in our discharges. Model simulations for these species are displayed as lines in figure 2. The agreement with the experimental results is good, reproducing the behaviour of the three molecules reasonably well. Water is shown to have a maximum concentration of ~ 35%, which is slightly higher than the experimental value (~ 30%) and corresponds also to the 40% O2 mixture. This slight overestimation of water formation happens for all the simulated conditions, leading to marginally higher depletions of the precursors H2 and O2 than those measured, especially for the extreme mixtures. Steady state concentrations of H2O are given by the equilibrium between the dissociation of the precursor gases and the recombination of radicals at the wall. These radicals have not
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 7
Europe PMC Funders Author Manuscripts
been measured experimentally, but their concentrations can be simulated with the model. Figure 3 shows the predicted relative concentrations of all the neutral species considered, normalized to the sum of concentrations of all of them. As could be expected, the stable neutral species dominate the distributions, but radicals appear in considerable amounts. Atomic hydrogen, produced mainly through dissociation of H2, has concentrations of ~ 5% for most of the mixtures, decreasing below this value only for mixtures containing < 10% H2. Oxygen atoms in the ground state show comparable concentrations to those of atomic hydrogen, but reach higher densities for O2-rich mixtures (up to 30% O(3P) for pure oxygen) and tend to decrease with growing H2 content. The other main radical, OH, is formed mainly from the dissociation of water molecules and thus follows a similar behaviour to this species, reaching concentrations up to 10% for high oxygen conditions. Ozone and HO2 are hardly formed in the plasma and their relevance to the chemistry is limited. The two excited species included in the model, O2(a1Δg) and O(1D), have very different effects in the chemistry of the discharge. Concentrations of O2(a1Δg) are comparatively high, typically ~ 5% of the O2 concentration, almost reaching 10% of the total neutral concentration for the pure oxygen plasma. Despite these high values, the impact in the chemistry of the discharge is limited, its main role being the production of O− ions through dissociative attachment. On the other hand, the amounts of O(1D) in the discharge are very low, three or four orders of magnitude below the concentration of O(3P), but with a heavier role in the neutral chemistry of the discharge. The reactions of O(1D) with H2 and H2O (G11 and G15) have high rate coefficients for a hom*ogeneous neutral process, and become a main source for OH radicals when the concentration of O(1D) is high enough (80% - 90% O2 mixtures).
Europe PMC Funders Author Manuscripts
Figure 4 shows the measured positive ion distributions for the different considered mixtures, as well as their model predictions. Due to the high density of data, the different species have been arranged in two panels for clarity. Hydrogenic ions (H+, H2+, H3+) are only dominant for mixtures with nearly no oxygen, where H3+ is clearly the dominant ion. As soon as oxygen is added to the mixture, water is formed and H3O+ quickly becomes the major ion, while the concentration of H3+ decreases. This situation is maintained through mixtures with approximately 4% < O2 < 70%, with H3O+ accounting for ~ 50-60 % of the positive charge. In this range, the pure oxygenic ions O+ and O2+ increase with the O2 fraction, while mixed ions (H2O+, OH+, HO2+) are roughly stable. At ~ 80% O2 fraction, O2+ becomes the major ion, with H3O+ sinking, following the decrease in neutral H2O available in the gas phase. Between 90% and 100% O2, all hydrogen-containing ions abruptly disappear from the plasma as expected, with only O2+ and O+ in a 2:1 ratio remaining in the pure oxygen discharge. The global behaviour of the major ions is reasonably well reproduced by the model, although there are some discrepancies with the experimental observations. At the lowest O2 concentration, H3+ is the dominant ion, with an important contribution of H3O+ and H+, which replicates the experimental results. However, the value for the electron temperature used for this simulation is 2.4 eV, as opposed to the 1.7 eV value measured with the Langmuir probe (see Fig. 1). Using the measured value as an input for the model leads to an
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 8
important overestimation of the H3O+ concentration. The use of a higher value for the electron temperature is supported by previous measurements in pure hydrogen plasmas at the same pressure [33, 37, 38]. The rest of the model simulations have been carried out using the experimental values for the electron temperatures and densities.
Europe PMC Funders Author Manuscripts
The addition of oxygen to the mixture makes H3O+ quickly become the major ion. In H2rich plasmas, the main mechanism for the formation of this ion is the proton transfer from H3+ to H2O (reaction T12), whose high rate coefficient can be justified by taking into account the difference in proton affinity of the species involved (691.0 kJ/mol for H2O and 422.3 kJ/mol for H2). As the oxygen ratio grows, other ions become important in the mixture while H3+ sinks, diversifying the production processes for H3O+. For intermediate mixtures, these include charge transfer reactions involving H2O+ (reactions T21 and T22) and OH+ (T19), all of which have high rate coefficients. When the hydrogen present in the mixture is lowered, the concentration of hydronium decreases and O2+, which is mainly produced from the direct ionization of O2 molecules, becomes the major ion. For the pure oxygen plasma, only this ion and O+ from the dissociative ionization of O2 remain, but the ratio between them (3:1) is different from the experimental one (2:1). This ratio is highly dependent on the electron temperature, so a rise in this parameter would help reproduce the experimental results.
Europe PMC Funders Author Manuscripts
The three mixed ions OH+, H2O+ and HO2+ have concentrations of the order of ~ 1-3% for most of the mixtures studied with H2O+ tending to be slightly more abundant. This behaviour reproduces qualitatively the experimental observations. H2O+ is produced mainly through the ionization of water molecules (I8), and then destroyed through collisions with H2 (T21) and H2O (T22), as mentioned earlier. In the case of OH+, it is produced mainly from the dissociative ionization of H2O (I9), and destroyed through collisions with all of the major molecular species in the plasma, H2 (T17), O2 (T20) and H2O (T18 and T19). For HO2+ , the reaction with H2 to form H3+ is nearly thermoneutral, with the forward (T25) and backward (T13) rate coefficients being comparable, and this equilibrium determines its concentration for H2-rich mixtures. When oxygen is the major component of the mixture, reaction T24 becomes the main source of HO2+, while the destruction mechanism is still the same. It is interesting to observe at this point that the relative concentrations of the H3O+, H2O+ and OH+ ions are similar to those predicted in astrochemical models for the interior of dark clouds [8], where the chemistry leading to the formation of these ions has some similitude with that of our discharge. It starts also with the ionization of molecular hydrogen and proceeds through proton transfer reactions involving H3+. In diffuse clouds, where OH+ and H2O+ have been detected [6, 7], H2 is scarce, and the abundant electrons neutralize H3+. In these regions other mechanisms starting with the ionization of atomic hydrogen become prevalent and produce an inverse ordering of the ionic relative abundances [8] ([OH+] > [H2O+] > [H3O+]). The concentration of the O2H+ ion found in our plasmas is similar to that of OH+, which suggests that this ion might be found, as a minor species, inside dark clouds. Purely hydrogenic ions obviously dominate for pure H2 conditions but decrease abruptly with the addition of oxygen to the mixture, especially H3+, falling orders of magnitude from
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 9
their initial values. The experimental data shows the same decrease but with a smoother slope, with H3+ only going below 1% concentration for a ~ 60% O2 mixture, while in the model simulations it occurs for a mere 10% O2. In the case of H2+, its concentration is underestimated by almost an order of magnitude through the whole range of mixtures, and for H+ the behaviour is only reproduced for high H2 and high O2 conditions.
Europe PMC Funders Author Manuscripts
The concentrations of these ions depend strongly on the electron temperature, as both H+ and H2+ are produced by electron impact ionization of H2 (reactions I5 and I6, respectively), and H3+ is produced from H2+ in a charge transfer reaction with H2 (T5). A higher electron temperature would enhance the production of these ions, leading to a better agreement between the experiment and the model. However, for the majority of mixtures studied, rising the electron temperature would lead to an important increase in certain species, such as H2O+ and OH+, which are already well reproduced by the model, effectively altering the behaviour of the major ions in the discharge. Figure 5 displays the simulated relative concentrations of positive ions for two electron temperatures: 2 eV and 3 eV. The variations commented on above can be appreciated.
Europe PMC Funders Author Manuscripts
The results of the simulations for the negative charge carriers can be seen in figure 6. The distribution is dominated by electrons for all mixture ratios, with negative ions making up at most 25% of the total negative charge. In the pure H2 plasma, negative ions are hardly present, with a relative concentration of H− ~ 0.1%. However, the addition of small amounts of O2 to the mixture makes its concentration rise up to 10%, due to the high cross section for the dissociative attachment of H2O (reaction A2), which is being formed in the discharge. As the oxygen ratio grows, H− ions are lost due to collisional detachment with O2 (Dt4), whereas O− and OH- ions increase in concentration, O− being produced from the dissociative attachment of water (A5) and then transformed into OH− through collisions with H2 and H2O (T26 and T27). OH− is the major negative ion for mixtures from 40% to 80% O2. For higher oxygen concentrations, the amount of H2 and H2O in the discharge is not enough to maintain an efficient formation of OH− , so it decreases quickly. In these conditions, O− is formed from the dissociative attachment of O2 (reaction A1) and O2(a1Δg) (reaction A4). The relevance of the negative ion processes in the global chemistry of the discharge is low, as they are not involved in any main mechanism for production or destruction of positive ions or neutrals. In our discharge, ion-ion neutralization, which is the main gas phase process involving positive ions, is far less important than the neutralization at the cathode walls. The main contribution of negative ions is in fact the decrease in the electron density, which lowers the rate of electron impact processes but, given that their concentration never exceeds 25% of the total negative charge, this effect is not large.
5. Summary and conclusions A combined diagnostics and modelling of low pressure H2 + O2 plasmas with different mixture ratios, generated in a hollow cathode DC reactor, has been presented. The results of the model simulations have allowed the identification of the main processes determining the observed neutral and ion distributions. To our knowledge, no combined experimental and
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 10
modelling studies of the kinetics of low pressure H2 + O2 plasmas including ions had been reported before.
Europe PMC Funders Author Manuscripts
Water formation is observed in the discharge for the whole range of mixtures, up to a maximum of ~ 30% relative concentration. These results are well reproduced by the kinetic model, which predicts slightly higher concentrations (up to 35%) of water. The concentrations of other neutrals, including radicals and excited species that could not be experimentally observed, have been also simulated with the model. Atomic oxygen and hydrogen are formed in appreciable amounts, with relative concentrations of the order of 10% when the presence of their molecular precursors is significant. The OH radical is formed from the dissociation of water, showing a similar behaviour over the range of mixtures, with a peak value of ~ 10% of the total neutral concentration. The excited species O2(a1Δg) and O(1D) are formed in different proportions in the discharge, with O2(a1Δg) reaching up to 10% relative concentration but having a limited impact in the chemistry. In contrast, O(1D) is produced in smaller amounts (up to ~ 0.1%) but has a great relevance in the chemistry due to the high cross sections for its reactions with neutrals. The other neutral species considered, O3 and HO2, are hardly formed in our plasmas.
Europe PMC Funders Author Manuscripts
The experimental ion distributions are dominated by hydrogenic ions only for mixtures with nearly no oxygen. As soon as H2O is formed in the plasma, H3O+ becomes the major ion, remaining as such for mixtures with 4% < O2 < 70%. For higher O2 fractions, pure oxygen ions become dominant, with O2+ as the major ion for these mixtures. The concentrations of the mixed minor ions H2O+, OH+, HO2+ are stable through a wide range of H2/O2 ratios, only sinking for the extreme mixtures. A comparison of the relative abundances of these ions with the predictions of astrochemical models suggests that HO2+ might be present as a minor component in the interior of dark interstellar clouds. Model simulations reproduce the behaviour of the ions reasonably well. For the pure H2 discharge, H3+ is the dominant ion, and when oxygen is added to the mixture, H3O+ concentration grows due to proton transfer between H3+ and H2O, becoming the major ion. For high O2 ratios, direct ionization of this precursor causes O2+ to prevail in the plasma. The main discrepancy between measurements and simulations is found for the pure hydrogenic ions, whose predicted decrease upon oxygen addition to the mixture is much more abrupt than experimentally observed. This is due to the low electron temperature, which causes the charge transfer processes to prevail over electron impact. As soon as oxygen is added to the mixture, water is formed and H3+ is destroyed through charge transfer to H2O. A higher value for the electron temperature would increase the concentration of hydrogenic ions through the direct ionization of H2 and subsequent charge transfer; however, this would lead to a great change in the concentrations of other ions. A non-maxwellian electron energy distribution function could justify a different balance of these processes, by increasing or decreasing the population of the high energy tail. The concentrations of negative charge carriers have been simulated with the model. The distribution is dominated by electrons for all mixture ratios, with negative ion concentrations reaching up to 25%. H− and O− are the major negative ions for the H-rich and O-rich discharges, respectively, while OH− prevails for the intermediate mixtures. The relevance of
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 11
negative ions to the global chemistry is limited. Their main function is decreasing the electron density available for electron impact processes.
Acknowledgements Europe PMC Funders Author Manuscripts
This work has been funded by the MCINN of Spain under grants FIS2010-16455, FIS2013-48087-C2-1-P and the Consolider Astromol, CDS2009-00038. We thank also the European Research Council for additional support under ERC-2013-Syg 610256-NANOCOSMOS. MJR has received funding from the FPI program of the MCINN, and EC from the JdC program of the same ministry. We are indebted to M. A. Moreno and J. Rodríguez for technical assistance.
References
Europe PMC Funders Author Manuscripts
[1]. Herbst E. Philos. Trans. R. Soc. A-Math. Phys. Eng. Sci. 2000; 358:2523–33. [2]. Herbst E. Chem. Soc. Rev. 2001; 30:168–76. [3]. Meyer, DM.; van Dishoeck, EF. Molecules in Astrophysics: Probes and Processes. Springer; Dordrecht: 1997. p. 407-19. [4]. Williams TL, Adams NG, Babco*ck LM, Herd CR, Geoghegan M. Mon. Not. Roy. Astron. Soc. 1996; 282:413–20. [5]. Wootten A, Mangum JG, Turner BE, Bogey M, Boulanger F, Combes F, Encrenaz PJ, Gerin M. Astrophys. J. 1991; 380:L79–L83. [6]. Wyrowski F, Menten KM, Güsten R, Belloche A. A&A. 2010; 518:A26. [7]. Ossenkopf V, et al. A&A. 2010; 518:L111. [8]. Hollenbach D, Kaufman MJ, Neufeld D, Wolfire M, Goicoechea JR. Astrophys. J. 2012; 754:105. [9]. Herbst E, Green S, Thaddeus P, Klemperer W. Astrophys. J. 1977; 215:503–10. [10]. Weaver SLW, Woon DE, Ruscic B, McCall BJ. Astrophys. J. 2009; 697:601–9. [11]. Kluge L, Gartner S, Brunken S, Asvany O, Gerlich D, Schlemmer S. Philos. Trans. R. Soc. AMath. Phys. Eng. Sci. 2012; 370:5041–54. [12]. Mushiaki M. Fusion Eng. Des. 1999; 47:85–97. [13]. Waelbroeck F, Winter J, Wienhold P. J. Vac. Sci. Technol. A. 1984; 2:1521–36. [14]. Counsell G, et al. Plasma Phys. Control. Fusion. 2006; 48:B189–B99. [15]. Tabarés FL, Ferreira JA, Tafalla D, Tanarro I, Herrero VJ, Méndez I, Gómez-Aleixandre C, Albella JM. J. Phys.: Conf. Ser. 2008; 100:062025. [16]. Ceccone G, Gilliland D, Kylian O, Rossi F. Czech. J. Phys. 2006; 56:B672–B7. [17]. Huang C, Tsai CY, Juang RS. J. Appl. Polym. Sci. 2012; 124:E108–E15. [18]. Rybkin V, Kuvaldina E, Grinevich A, Choukourov A, Iwai H, Biederman H. Plasma Process. Polym. 2008; 5:778–87. [19]. Fumagalli F, Kylian O, Amato L, Hanus J, Rossi F. J. Phys. D-Appl. Phys. 2012; 45:135203. [20]. Thandavan T, Wong C, Gani S, Nor R. Appl. Nanosci. 2012; 2:47–53. [21]. Konstantinovskii RS, Shibkov VM, Shibkova LV. Kinet. Catal. 2005; 46:775–88. [22]. Breden D, Raja LL. IEEE Trans. Plasma Sci. 2011; 39:2250–1. [23]. Peyrous R, Pignolet P, Held B. J. Phys. D-Appl. Phys. 1989; 22:1658–67. [24]. Vasko CA, Liu DX, van Veldhuizen EM, Iza F, Bruggeman PJ. Plasma Chem. Plasma Process. 2014; 34:1081–99. [25]. Liu DX, Bruggeman P, Iza F, Rong MZ, Kong MG. Plasma Sources Sci. Technol. 2010; 19:22. [26]. Yin Z, Adamovich IV, Lempert WR. Proc. Combust. Inst. 2013; 34:3249–58. [27]. Ho WC, Pursell CJ, Oka T. J. Mol. Spectrosc. 1991; 149:530–41. [28]. Czerwiec T, Gavillet J, Belmonte T, Michel H, Ricard A. J. Phys. III. 1996; 6:1205–12. [29]. Levanov AV, Antipenko EE, Lunin VV. Moscow Univ. Chem. Bull. 2010; 65:341–9. [30]. Du KY, Krogh O, Setser DW. Int. J. Chem. Kinet. 2006; 38:12–7. [31]. Tanarro I, Herrero VJ. Plasma Sources Sci. Technol. 2011; 20
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 12
Europe PMC Funders Author Manuscripts Europe PMC Funders Author Manuscripts
[32]. Carrasco E, Jiménez-Redondo M, Tanarro I, Herrero VJ. Plasma Phys. Control. Fusion. 2012; 54:124019. [33]. Mendez I, Gordillo-Vázquez FJ, Herrero VJ, Tanarro I. J. Phys. Chem. A. 2006; 110:6060–6. [PubMed: 16671676] [34]. Castillo M, Mendez I, Islyaikin AM, Herrero VJ, Tanarro I. J. Phys. Chem. A. 2005; 109:6255– 63. [PubMed: 16833966] [35]. de los Arcos T, Domingo C, Herrero VJ, Sanz MM, Schulz A, Tanarro I. J. Phys. Chem. A. 1998; 102:6282–91. [36]. Castillo M, Herrero VJ, Tanarro I. Plasma Sources Sci. Technol. 2002; 11:368–76. [37]. Jiménez-Redondo M, Carrasco E, Herrero VJ, Tanarro I. Phys. Chem. Chem. Phys. 2011; 13:9655–66. [PubMed: 21491042] [38]. Carrasco E, Jiménez-Redondo M, Tanarro I, Herrero VJ. Phys. Chem. Chem. Phys. 2011; 13:19561–72. [PubMed: 21984202] [39]. Carrasco E, Tanarro I, Herrero VJ, Cernicharo J. Phys. Chem. Chem. Phys. 2013; 15:1699–706. [PubMed: 23247609] [40]. Jiménez-Redondo M, Cueto M, Doménech JL, Tanarro I, Herrero VJ. RSC Adv. 2014 DOI: 10.1039/C4RA13102A. [41]. Kim YK, Desclaux JP. Phys. Rev. A. 2002; 66:012708. [42]. Itikawa Y. J. Phys. Chem. Ref. Data. 2009; 38:1–20. [43]. Tarnovsky V, Deutsch H, Becker K. J. Chem. Phys. 1998; 109:932–6. [44]. Itikawa Y, Mason N. J. Phys. Chem. Ref. Data. 2005; 34:1–22. [45]. Gudmundsson JT, Kouznetsov IG, Patel KK, Lieberman MA. J. Phys. DAppl. Phys. 2001; 34:1100–9. [46]. Lee C, Graves DB, Lieberman MA, Hess DW. J. Electrochem. Soc. 1994; 141:1546–55. [47]. van Dishoeck EF, Langhoff SR, Dalgarno A. J. Chem. Phys. 1983; 78:4552–61. [48]. Peverall R, et al. J. Chem. Phys. 2001; 114:6679–89. [49]. Mitchell JBA. Phys. Rep.-Rev. Sec. Phys. Lett. 1990; 186:215–48. [50]. Rosen S, et al. Faraday Discuss. 2000; 115:295–302. [PubMed: 11040515] [51]. Jensen MJ, Bilodeau RC, Safvan CP, Seiersen K, Andersen LH, Pedersen HB, Heber O. Astrophys. J. 2000; 543:764–74. [52]. Woodall J, Agúndez M, Markwick-Kemper AJ, Millar TJ. A&A. 2007; 466:1197–204. [53]. Neyts E, Yan M, Bogaerts A, Gijbels R. J. Appl. Phys. 2003; 93:5025–33. [54]. Gordillo-Vázquez FJ. J. Phys. D-Appl. Phys. 2008; 41:234016. [55]. Eichwald O, Yousfi M, Hennad A, Benabdessadok MD. J. Appl. Phys. 1997; 82:4781–94. [56]. Kossyi IA, Kostinsky AY, Matveyev AA, Silakov VP. Plasma Sources Sci. Technol. 1992; 1:207. [57]. Stafford DS, Kushner MJ. J. Appl. Phys. 2004; 96:2451–65. [58]. Atkinson R, Baulch DL, Cox RA, Hampson RF, Kerr JA, Rossi MJ, Troe J. J. Phys. Chem. Ref. Data. 1997; 26:1329–499. [59]. Anicich VG. J. Phys. Chem. Ref. Data. 1993; 22:1469–569. [60]. Millar TJ, Farquhar PRA, Willacy K. Astron. Astrophys. Suppl. Ser. 1997; 121:139–85. [61]. Ferguson EE. At. Data Nucl. Data Tables. 1973; 12:159–78. [62]. Jaffke T, Meinke M, Hashemi R, Christophorou LG, Illenberger E. Chem. Phys. Lett. 1992; 193:62–8. [63]. Tawara H, Itikawa Y, Nishimura H, Yoshino M. J. Phys. Chem. Ref. Data. 1990; 19:617–37. [64]. Janev, RK.; Langer, WD.; Evans, JK.; Post, DE. Elementary Processes in Hydrogen - Helium Plasmas: Cross Sections and Reaction Rate Coefficients. Springer; Berlin: 1987. [65]. Slanger TG, Black G. J. Chem. Phys. 1978; 68:989–97. [66]. Dunkin DB, Fehsenfeld FC, Ferguson EE. J. Chem. Phys. 1970; 53:987–9. [67]. McEwan, MJ.; Phillips, LF. Chemistry of the atmosphere. Halsted Press; New York: 1975. [68]. Riahi R, Teulet P, Ben Lakhdar Z, Gleizes A. Eur. Phys. J. D. 2006; 40:223–30.
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 13
Europe PMC Funders Author Manuscripts
[69]. Belostotsky SG, Economou DJ, Lopaev DV, Rakhimova TV. Plasma Sources Sci. Technol. 2005; 14:532–42. [70]. McFarland M, Albritton DL, Fehsenfeld FC, Ferguson EE, Schmeltekopf AL. J. Chem. Phys. 1973; 59:6629–35. [71]. Scott CD, Farhat S, Gicquel A, Hassouni K. J. Thermophys. Heat Transfer. 1996; 10:426–35. [72]. Hassouni K, Grotjohn TA, Gicquel A. J. Appl. Phys. 1999; 86:134–51. [73]. Gudmundsson JT, Thorsteinsson EG. Plasma Sources Sci. Technol. 2007; 16:399–412. [74]. Sharpless RL, Slanger TG. J. Chem. Phys. 1989; 91:7947–50. [75]. Haas FA, Goodyear A, Braithwaite NSJ. Plasma Sources Sci. Technol. 1998; 7:471–7. [76]. Gordillo-Vázquez FJ, Donkó Z. Plasma Sources Sci. Technol. 2009; 18 [77]. Boffard JB, Jung RO, Lin CC, Aneskavich LE, Wendt AE. Plasma Sources Sci. Technol. 2011; 20 [78]. Gill P, Webb CE. J. Phys. D-Appl. Phys. 1977; 10:299–301. [79]. Bogaerts A, Gijbels R, Goedheer WJ. J. Appl. Phys. 1995; 78:2233–41. [80]. Mendez I, Tanarro I, Herrero VJ. Phys. Chem. Chem. Phys. 2010; 12:4239–45. [PubMed: 20379518] [81]. Jiménez-Redondo M, Carrasco E, Herrero VJ, Tanarro I. Plasma Sources Sci. Technol. 2013; 22:025022. [82]. Herrero VJ, Islyaikin AM, Tanarro I. J. Mass Spectrom. 2008; 43:1148–50. [PubMed: 18300331] [83]. Tanarro I, Herrero VJ. Plasma Sources Sci. Technol. 2009; 18:034007. [84]. Parra-Rojas FC, Passas M, Carrasco E, Luque A, Tanarro I, Simek M, Gordillo-Vázquez FJ. J. Geophys. Res-Space Phys. 2013; 118:4649–61. [85]. Gousset G, Touzeau M, Vialle M, Ferreira CM. Plasma Chem. Plasma Process. 1989; 9:189–206. [86]. Singh H, Coburn JW, Graves DB. J. Appl. Phys. 2000; 88:3748–55. [87]. Hodgson A, Wight A, Worthy G. Surf. Sci. 1994; 319:119–30. [88]. Błoński P, Kiejna A, Hafner J. Phys. Rev. B. 2008; 77:155424. [89]. Govender A, Curulla Ferré D, Niemantsverdriet JW. ChemPhysChem. 2012; 13:1583–90. [PubMed: 22298316] [90]. Jung SC, Kang MH. Phys. Rev. B. 2010; 81:115460.
Europe PMC Funders Author Manuscripts Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 14
Europe PMC Funders Author Manuscripts Europe PMC Funders Author Manuscripts
Figure 1.
Electron temperature (a) and electron density (b) measured with a double Langmuir probe. The values used in the model are the same as the experimental ones except for the pure H2 discharge, where an electron temperature of 2.4 eV is employed (see text).
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 15
Europe PMC Funders Author Manuscripts Europe PMC Funders Author Manuscripts
Figure 2.
Relative concentrations for the stable neutrals as a function of O2 fraction, normalized to the sum of concentrations of the three stable species. Symbols: experimental data. Lines: values obtained from model simulations.
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 16
Europe PMC Funders Author Manuscripts Europe PMC Funders Author Manuscripts
Figure 3.
Simulated relative concentrations for all the neutral species considered in the model, normalized to the sum of concentrations of all of them. They are split in two panels for clarity.
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 17
Europe PMC Funders Author Manuscripts Europe PMC Funders Author Manuscripts
Figure 4.
Symbols: experimental relative concentrations of positive ions in the discharge for the different mixtures studied, normalized to the sum of concentrations of all positive ions. The narrow dotted lines are only to guide the eye. Continuous lines: model results for the relative concentrations of positive ions. The data have been split in two panels for clarity.
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 18
Europe PMC Funders Author Manuscripts Europe PMC Funders Author Manuscripts
Figure 5.
Model predictions of relative concentrations of positive ions in the discharge at two electron temperatures, 2 eV and 3 eV, normalized to the sum of concentrations of all positive ions.
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 19
Europe PMC Funders Author Manuscripts Europe PMC Funders Author Manuscripts
Figure 6.
Model simulations for relative concentrations of negative charge carrier normalized to the sum of concentrations of all of them.
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 20
Table 1
Gas phase reactions included in the model.
Europe PMC Funders Author Manuscripts
Process
Rate coefficient (cm3 s−1)
Reference
Electron impact ionization I1
e + O → O+ + 2e
1.03 × 10−8 Te0.5 exp(−14.3/Te)
[41]
I2
e + O2 → O+ + O + 2e
4.84 × 10−9 Te0.5 exp(−22.5/Te)
[42]
I3
e + O2 → O2+ + 2e
7.07 × 10−9 Te0.5 exp(−13.1/Te)
[42]
I4
e + H → H+ + 2e
6.50 × 10−9 Te0.49 exp(−12.89/Te)
[33]
I5
e + H2 → H+ + H + 2e
3.00 × 10−8 Te0.44 exp(−37.72/Te)
[33]
I6
e + H2 → H2+ + 2e
3.12 × 10−8 Te0.17 exp(−20.07/Te)
[33]
I7
e + OH → OH+ + 2e
1.48 × 10−8 Te0.5 exp(−12.6/Te)
[43]
I8
e + H2O → H2O+ + 2e
9.87 × 10−9 Te0.5 exp(−13.3/Te)
[44]
I9
e + H2O → OH+ + H + 2e
2.88 × 10−9 Te0.5 exp(−17.7/Te)
[44]
I10
e + H2O → H+ + OH + 2e
1.77 × 10−9 Te0.5 exp(−20.0/Te)
[44]
I11
e + H2O → O+ + H2 + 2e
3.03 × 10−10 Te0.5 exp(−23.5/Te)
[44]
I12
e + O2(a) → O2+ + 2e
9.0 × 10−10 Te2.0 exp(−11.6/Te)
[45]
Electron impact dissociation
Europe PMC Funders Author Manuscripts
D1
e + H2 → 2H + e
1.75 × 10−7 Te−1.24 exp(−12.59/Te)
[33]
D2
e + O2 → 2O + e
4.2 × 10−9 exp(−5.56/Te)
[46]
D3
e + O2 → O + O1D + e
5.0 × 10−8 exp(−8.40/Te)
[46]
D4
e + OH → O + H + e
KD4
D5
e + H2O → OH + H + e
D6 D7
a
[47]
KD5
b
[44]
e + H2O → O1D + H2 + e
2.0 × 10−9 Te0.5 exp(−7.0/Te)
[44]
e + O2(a) → 2O + e
4.2 × 10−9 exp(−4.6/Te)
[45]
Electron impact neutralization c
N1
H2+ + e → H + H
KN1
N2
H3+ + e → 3H
0.5 × KN2
N3
H3+ + e → H2 + H
N4
O2+ + e → O + O
N5
+
O2 + e → O + +
N6
O2 + e →
N7
OH+
N8
O+
N9
+
O1D
+e→O+H + e → OH + H
H2
O+
H2
O1D
O1D
+ e → O + H2
[33] d
[33]
0.5 × KN2
d
[33]
4.9 × 10−8 (0.026/Te)0.7
[48]
1.06 ×
10−7
(0.026/Te)0.7
[48]
7.56 ×
10−8
(0.026/Te)0.7
[48]
3.75 ×
10−8
(0.026/Te)0.5
[49]
8.6 ×
10−8
)0.5
[50]
3.9 ×
10−8
)0.5
[50]
(0.026/Te (0.026/Te
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 21
Process
Rate coefficient (cm3 s−1)
Reference
Electron impact ionization
Europe PMC Funders Author Manuscripts
N10
H2O+ + e → O + H + H
3.05 × 10−7 (0.026/Te)0.5
[50]
N11
H3O+ + e → OH + H +H
2.85 × 10−7 (0.026/Te)0.5
[51]
N12
H3O+ + e → O + H2 + H
5.6 × 10−9 (0.026/Te)0.5
[51]
N13
H3O+ + e → OH + H2
6.02 × 10−8 (0.026/Te)0.5
[51]
N14
H3O+ + e → H2O + H
1.08 × 10−7 (0.026/Te)0.5
[51]
N15
HO2+ + e → O2 + H
3 × 10−7 (0.026/Te)0.5
[52]
Neutral hom*ogeneous
Europe PMC Funders Author Manuscripts
G1
H + O3 → O + HO2
7.51 × 10−13
[53]
G2
H + HO2 → H2O + O
9.18 × 10−11 exp(−971.9/Tg)
[54]
G3
H + HO2 → H2O + O(1D)
4.8 × 10−16 Tg1.55 exp(80.58/Tg)
[53]
G4
H + HO2 → O2 + H2
1.1 × 10−12 Tg0.56 exp(−346/Tg)
[54]
G5
H + HO2 → 2OH
2.35 × 10−10 exp(−373.7/Tg)
[54]
G6
O(1D) + HO2 → OH + O2
2.9 × 10−11 exp(200/Tg)
[54]
G7
O2(a) + HO2 → OH + O + O2
1.66 × 10−11
[55]
G8
H + O3 → OH + O2
2.71 × 10−11 (Tg/300)0.75
[53]
G9
O(1D) + O3 → 2O2
1.2 × 10−10
[53]
G10
O(1D) + O3 → 2O + O2
1.2 × 10−10
[53]
G11
O(1D) + H2 → OH + H
1.1 × 10−10
[54]
G12
O(1D) + O2 → O + O2
4.8 × 10−12 exp(67/Tg)
[56]
G13
O(1D) + O2 → O + O2(a)
1.6 × 10−12 exp(67/Tg)
[57]
G14
O(1D) + OH → H + O2
6 × 10−11 Tg−0.186 exp(−154/Tg)
[54]
G15
O(1D) + H2O → 2OH
1.62 × 10−10 exp(64.95/Tg)
[53]
G16
O(1D) + H2O → O + H2O
1.2 × 10−11
[58]
3.75 × 10−10
[59]
8.20 × 10−9
[59]
Ion-molecule T1 T2 T3 T4
H + + O → O+ + H H+
O+
+ H2O → H2
H+
+H
+
+ O2 → O2 + H
1.17 × 10−9
[59]
H+
6.40 × 10−10
[59]
+H
2.00 × 10−9
[59]
+
H2 + H → H2 + +
H3+
T5
H2 + H2 →
T6
+
O+
+ H2
0.53 × 7.30 × 10−9 = 3.87 × 10−9
[59]
+
H3O+
+H
0.47 × 7.30 × 10−9 = 3.43 × 10−9
[59]
T7 T8 T9 T10
H2 + H2O → H2 H2 + H2O → +
+
H2 + O2 → O2 + H2 +
0.29 × 2.70 × 10−9 = 7.83 × 10−10
[59]
+
0.71 × 2.70 × 10−9 = 1.92 × 10−9
[59]
+ H2
0.70 × 1.20 × 10−9 = 8.40 × 10−10
[59]
H2 + O2 → HO2 + H +
H3 + O →
OH+
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Process
Page 22
Rate coefficient (cm3 s−1)
Reference
Electron impact ionization
Europe PMC Funders Author Manuscripts
T11
H3+ + O → H2O+ + H
0.30 × 1.20 × 10−9 = 3.60 × 10−10
[59]
T12
H3+ + H2O → H3O+ + H2
5.30 × 10−9
[59]
T13
H3+ + O2 → HO2+ + H2
6.70 × 10−10
[59]
T14
O + + H → H+ + O
6.40 × 10−10
[59]
T15
O+ + H2 → OH+ + H
1.62 × 10−9
[59]
T16
O+ + H2O → H2O+ + O
2.60 × 10−9
[59]
T17
OH+ + H2 → H2O+ + H
9.70 × 10−10
[59]
T18
OH+ + H2O → H2O+ + OH
0.55 × 2.89 × 10−9 = 1.59 × 10−9
[59]
T19
OH+ + H2O → H3O+ + O
0.45 × 2.89 × 10−9 = 1.30 × 10−9
[59]
T20
OH+ + O2 → O2+ + OH
3.80 × 10−10
[59]
T21
H2O+ + H2 → H3O+ + H
7.60 × 10−10
[59]
T22
H2O+ + H2O → H3O+ + OH
1.85 × 10−9
[59]
T23
H2O+ + O2 → O2+ + H2O
3.30 × 10−10
[59]
T24
O2+ + H2 → HO2+ + H
4.00 × 10−11
[59]
T25
HO2+ + H2 → H3+ + O2
3.30 × 10−10
[59]
T26
O− + H2 → OH− + H
3 × 10−11
[60]
T27
O− + H2O → OH− + OH
1.4 × 10−9
[61]
Electron impact attachment
Europe PMC Funders Author Manuscripts
A1
e + O2 → O− + O
1.07 × 10−9 Te−1.391 exp(−6.26/Te)
[62]
A2
e + H2O → OH + H−
3.54 × 10−9 Te−1.5 exp(−6.66/Te)
[44]
A3
e + H2 → H− + H
5.6 × 10−13 Te0.5 exp(−5.5/Te)
[63]
A4
e + O2(a) → O + O−
2.28 × 10−10 exp(−2.29/Te)
[45]
A5
e + H2O → H2 + O−
7.08 × 10−10 Te−1.3 exp(−8.61/Te)
[44]
A6
e + H2O → OH− + H
1.24 × 10−10 Te−1.3 exp(−7.32/Te)
[44]
Detachment Dt1
e + H− → H + 2e
2.32 × 10−8 Te2 exp(−0.13/Te)
[64]
Dt2
H− + H → H2 + e
1.3 × 10−9
[65]
Dt3
H− + O → OH + e
1 × 10−9
[65]
1.2 × 10−9
[66]
1.8 × 10−9
[67]
2 × 10−10
[67]
−1.9
[68]
Dt4
H−
Dt5
OH−
+ H → H2O + e
Dt6
OH−
+ O → HO2 + e
+ O2 → HO2 + e
Dt7
e + OH− → OH + 2e
Dt8
O−
Dt9
+ O2(a) → O3 + e
O−
+ H → OH + e
9.67 ×
10−6
Te
exp(−12.1/Te)
1.9 × 10−10 5×
10−10
[69] [60]
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Process
Page 23
Rate coefficient (cm3 s−1)
Reference
Electron impact ionization
Europe PMC Funders Author Manuscripts
Dt10
O− + H2 → H2O + e
6 × 10−10 (Tg/300)−0.24
[70]
Dt11
O− + O → O2 + e
2.3 × 10−10
[69]
Electron impact excitation and deexcitation X1
e + O2 → O2(a) + e
1.7 × 10−9 exp(−3.1/Te)
[45]
X2
e + O → O(1D) + e
4.5 × 10−9 exp(−2.29/Te)
[45]
Dx1
e + O2(a) → O2 + e
5.6 × 10−9 exp(−2.2/Te)
[45]
Ion-ion neutralization
Europe PMC Funders Author Manuscripts
IN1
H+ + H− → 2H
1.8 × 10−7 (Tg/300)−0.5
[21]
IN2
H2+ + H− → H + H2
2 × 10−7 (Tg/300)−0.5
[71]
IN3
H3+ + H− → 2H2
2 × 10−7 (Tg/300)−0.5
[72]
IN4
O+ + H− → H + O
2.3 × 10−7 (Tg/300)−0.5
[60]
IN5
O2+ + H− → H + O2
2 × 10−7 (Tg/300)−0.5
[56]
IN6
OH+ + H− → H2O
2 × 10−7 (Tg/300)−0.5
[56]
IN7
H2O+ + H− → H + H2O
2 × 10−7 (Tg/300)−0.5
[56]
IN8
H3O+ + H− → H2 + H2O
2.3 × 10−7 (Tg/300)−0.5
[60]
IN9
H+ + O− → H + O
2 × 10−7 (Tg/300)−0.5
[56]
IN10
H2+ + O− → H2O
2 × 10−7 (Tg/300)−0.5
[56]
IN11
H3+ + O− → OH + H2
2 × 10−7 (Tg/300)−0.5
[56]
IN12
O+ + O− → 2O
2 × 10−7 (Tg/300)−1
[57]
IN13
O2+ + O− → O2 + O
2 × 10−7 (Tg/300)−0.5
[73]
IN14
OH+ + O− → HO2
2 × 10−7 (Tg/300)−0.5
[56]
IN15
H2O+ + O− → O + H2O
2 × 10−7 (Tg/300)−0.5
[56]
IN16
H3O+ + O− → H2O + OH
2 × 10−7 (Tg/300)−0.5
[56]
IN17
H2+ + OH− → H2O + H
1 × 10−7
[56]
IN18
H3+ + OH− → H2 + H2O
2 × 10−7 (Tg/300)−0.5
[56]
IN19
O+ + OH− → HO2
2 × 10−7 (Tg/300)−0.5
[56]
IN20
O2+ + OH− → OH + O2
2 × 10−7 (Tg/300)−0.5
[56]
IN21
OH+ + OH− → 2OH
2 × 10−7 (Tg/300)−0.5
[56]
IN22
H2O+ + OH− → H2O + OH
2 × 10−7 (Tg/300)−0.5
[56]
IN23
H3O+ + OH− → 2H2O
4 × 10−7 (Tg/300)−0.5
[21]
Tg is given in K. Te is given in eV. a
KD4 = − 2.82402 × 10−11 Te + 3.38111 × 10−11 Te2 − 7.01504 × 10−12 Te3 + 6.09826 × 10−13 Te4 − 1.96671 × 10−14 Te5
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 24
b
KD5 = 1.67959 × 10−10 Te − 1.22568 × 10−11 Te2 + 2.19508 × 10−11 Te3 − 3.01892 × 10−12 Te4 + 1.2549 × 10−13 Te5
c KN1 = 7.51 × 10−9 − 1.12 × 10−9 Te + 1.03 × 10−10 Te2 − 4.15 × 10−12 Te3 + 5.86 × 10−14 Te4 d KN2 = 8.39 × 10−9 + 3.02 × 10−9 Te − 3.80 × 10−10 Te2 + 1.31 × 10−11 Te3 + 2.42 × 10−13 Te4 − 2.30 × 10−14 Te5 + 3.55 × 10−16 Te6
Europe PMC Funders Author Manuscripts Europe PMC Funders Author Manuscripts Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
Jiménez-Redondo et al.
Page 25
Table 2
Wall reactions for positive ions (neutralization) and neutrals (adsorption and recombination)
Europe PMC Funders Author Manuscripts
Wall neutralization
γ
K1
H+
K2
+
H2 + Wall→H2
1
K3
+
1
+ Wall→H
H3 + Wall→H2 + H
1
K4
O+
K5
+
O2 + Wall→O2
1
K6
OH+ + Wall→OH
1
K7
H2O+ + Wall→H2O
1
K8
H3O+ + Wall→H2O + H
1
K9
HO2+ + Wall→O2 + H
1
+ Wall→O
Heterogeneous reactions
1
γ
Ref.
Europe PMC Funders Author Manuscripts
W1
H + Fs→H(s)
1
e
W2
H(s) + H→H2 + Fs
0.0035
[39]
W3
O + Fs→O(s)
1
f
W4
O(s) + O→O2 + Fs
0.024
f
W5
OH + Fs→OH(s)
1
e
W6
H(s) + O→OH(s)
0.006
f
W7
H + O(s)→OH(s)
0.002
f
W8
OH(s) + H→H2O + Fs
0.004
f
W9
OH + H(s)→H2O + Fs
0.005
f
W10
O(s) + H2→H2O + Fs
0.00005
f
1
[57]
0.007
[74]
W11
O(1D)
W12
O2(a) + wall→O2
+ wall→O
Fs stands for a free surface site and X(s) refers to adsorbed species. e : Adsorption of atoms and radicals is assumed to have a probability of 1 f : Assumed in this work
Plasma Sources Sci Technol. Author manuscript; available in PMC 2015 December 21.
