Densities of Active Species in R/x%(N₂–5%H₂) (R = Ar or He) Microwave Flowing Afterglows

Abstract

Afterglows of R/x(N₂–5%H₂) (R = Ar or He) flowing microwave discharges are characterized by optical emission spectroscopy. Absolute densities of N atoms and N₂(A) and N₂(X,v > 13) molecules and estimated densities of NH and H atoms are determined after calibration of the N atom density by NO titration. It is determined the variations of active species densities along the post-discharge tube from the early afterglow to the late afterglow region. Conditions allowing to obtain a high concentration of H-atoms in dominant N-atoms have been found in the early afterglow region of the R/x(N₂–5%H₂) mixture for x < 5%. From these densities, the destruction probabilities of the H-atoms on the afterglow quartz tube wall is found to be: $\gamma_{H}^{He}$ = (2–3) × 10⁻⁴ and $\gamma_{H}^{Ar}$ = (3–4) × 10⁻³. The interest of these results concerns the enhancement of surface nitriding by combined effects of N and H atoms inclusion in afterglow conditions.

Introduction

Afterglows of N₂ flowing microwave discharges have previously been studied at medium gas pressures (1–20 Torr) for the sterilization of medical instruments by N-atoms [1, 2]. Recently [3, 4], it has been observed an enhanced nitriding of TiO₂ films in the flowing afterglow of N₂/2%H₂ microwave and RF plasmas, which was attributed to the presence of H and N atoms. Passivation of InGaN/GaN nanowires [5] and treatment of graphene films [6, 7] have also been studied in N₂ microwave afterglows.

In the present study, flowing afterglows produced by R/x%(N₂–5%H₂) (R = Ar or He) microwave plasmas have been studied by emission spectroscopy. Intensities emitted by the 11–7 band of the N₂ first positive system (1+) at 580 nm (I₅₈₀) and by the 1–0 band of the N₂ second positive system (2+) at 316 nm (I₃₁₆) were measured to obtain the absolute concentrations of N-atoms, N₂(A) and N₂(X,v > 13) after NO titration to calibrate the N-atom density. From these measurements, NH and H-atom densities have been estimated by choosing the appropriated kinetic reactions at the origin of the NH 336 nm emission.

Experimental Set-Up

The used experimental setup is reported in Fig. 1 [1, 2, 8,9,10].

Fig. 1

Photo and scheme of the microwave flowing afterglow experimental setup

Ar/N₂ and He/N₂ microwave discharges are produced by a surfatron cavity operating at 2.45 GHz. In these mixtures, the plasma length was found to extend between 2 and 20 cm after the surfatron gap, depending on the N₂ amount and on the HF power.

With a quartz discharge tube of 5 mm i.d. and 20 cm in length, connected to a bent post-discharge tube of 18 mm i.d. and 50 cm in length, the residence time of the afterglow at z = 3 cm after the beginning of the 18 mm post-discharge tube is 10⁻³ s.

A pre-mixed N₂–5%H₂ gas can be introduced instead of N₂ to produce NH radicals and H-atoms in addition to the N₂ active species.

In a previous work [11], first results were presented for the same R/x(N₂–5%H₂) (R = Ar or He) gas mixtures with x ≥ 20%. The present paper focuses on early afterglow conditions (z = 3 cm) and low N₂–5%H₂ percentages (x = 2–10%).

Constant operating conditions were used, with a total gas flow rate Q_total = 1.0 slpm, a pressure of 8 Torr and an injected microwave power P_MW = 150 W, that previously allow obtaining high concentrations of active species in the late afterglows of N₂/He and N₂/H₂ mixtures [11].

TiO₂ samples can eventually be exposed (heated or not) in a 5 l Pyrex reactor following the afterglow quartz tube (results not shown in this paper).

Optical emission spectroscopy (OES) along the afterglow tubes was performed using an optical fibre connected to an Acton Spectra Pro 2500i spectrometer (grating 600 g/mm), equipped with a Pixis 256E CCD detector (front illuminated 1024 × 256 pixels).

Active Species Densities in R/2-10%(N₂–5%H₂) Afterglows (R = Ar or He)

Table 1 gathers the kinetic reactions used in this work, with the corresponding rate coefficients and their origin in the literature.

Table 1 Kinetics reactions and corresponding rate coefficients in the R-N₂–H₂ afterglows

N-Atom, N₂(A), N₂ ⁺ and N₂(X, v > 13) Densities

In flowing afterglows, absolute N atom densities can be obtained by NO titration, a method extensively described in [8,9,10, 22] and well suited to the study of N₂ late afterglows between 5 and 10 Torr. In this pressure range, a good agreement has been found for absolute N-atom densities measured using the NO titration method and using the non-intrusive TALIF (Two-photon Absorption Laser-Induced Fluorescence) method [23].

In full late afterglow conditions, the N₂ (1+) (11–7) band emission at 580 nm (I₅₈₀) is produced by the 3-body recombination process of N atoms:

$${\text{N}}(^{4} {\text{S}}) \, + {\text{ N}}(^{4} {\text{S}}) \, + {\text{ M }} \to {\text{ N}}_{2} \left( {{\text{B}},{\text{v}}^{\prime } = 11} \right) \, + {\text{ M}}$$

(a)

$${\text{N}}_{2} ({\text{B}},{\text{v}}^{{\prime }} = 11) \, \to {\text{ N}}_{2} ({\text{A}},{\text{v}}^{{\prime \prime }} = 7) \, + {\text{ h}}\upnu \, \left( {1 + ,\;580\;{\text{nm}}} \right)$$

(a′)

In the early afterglow, here defined as the afterglow region lying between the pink and the late afterglow, only a fraction (a_N+N, with 0 ≤ a_N+N ≤ 1) of the I₅₈₀ emission is due to the process (a), the remaining fraction being caused by collisions between high vibrationally excited levels of the ground molecular state N₂(X,v > 13) and metastable states N₂(A):

$${\text{N}}_{2} ({\text{X}},{\text{v}} > 13) \, + {\text{ N}}_{2} ({\text{A}}) \, \to {\text{ N}}_{2} ({\text{B}},0 \le {\text{v}}^{{\prime \prime }} < 12) \, + {\text{ N}}_{2}$$

(b)

In consequence, in the early afterglow region and neglecting a possible quenching of the N₂(B,v′ = 11) level by H₂ molecules, the N atom concentration can be related to the a_N+N fraction of the observed I₅₈₀ emission through the proportionality relation:

$$a_{N + N} I_{580} = k_{1} [N]^{2}$$

(1)

with $k_{1} = c_{580} \frac{hc}{580}A_{{{\text{A}},7}}^{{{\text{B}},11}} \frac{{k_{a} [M]}}{{\left( {\upsilon_{{N_{2} \left( {{\text{B}},11} \right)}}^{R} + [R]k_{{N_{2} \left( {{\text{B}},11} \right)}}^{{Q_{R} }} + [N_{2} ]k_{{N_{2} \left( {{\text{B}},11} \right)}}^{{Q_{{N_{2} }} }} } \right)}}$ where $[M] = \frac{p}{kT}$ = 2.6 × 10¹⁷ cm⁻³ at p = 8 Torr and T = 300 K, $c_{580}$ is the spectral response of the spectral intensity acquisition system at 580 nm, $A_{{{\text{A}},7}}^{{{\text{B}},11}}$ is the vibrational transition probability of the N₂(1+, 11–7) band, $k_{a}$ is the rate coefficient for reaction (a), $\upsilon_{{N_{2} \left( {{\text{B}},11} \right)}}^{R}$ and $k_{{N_{2} \left( {{\text{B}},11} \right)}}^{{Q_{i} }}$ are respectively the radiative loss frequency and the quenching rates of the N₂(B,v″ = 11) level by species i (i = R or N₂).

As reported by Ricard et al. [11, 13], the N₂(A) and N₂(X,v > 13) densities can be deduced from the N-atom density by line ratio methods. Assuming that the following reation (c) is the dominant process for the production of the N₂(C,v′ = 1) level:

$${\text{N}}_{2} ({\text{A}}) \, + {\text{ N}}_{2} ({\text{A}}) \, \to {\text{ N}}_{2} ({\text{C}},{\text{v}}^{{\prime }} = 1) \, + {\text{ N}}_{2} ({\text{X}})$$

(c)

$${\text{N}}_{2} ({\text{C}},{\text{v}}^{{\prime }} = 1) \, \to {\text{ N}}_{2} ({\text{B}},{\text{v}}^{{\prime \prime }} = 0) \, + {\text{ h}}\upnu \, \left( {2 + ,\;316\;{\text{nm}}} \right)$$

(c′)

Using the same definitions than above, it comes for the $I_{316}$ intensity:

$$I_{316} = k_{2} \left[ {N_{2} (A)} \right]^{2}$$

(2)

with $k_{2} = c_{316} \frac{hc}{316}A_{{{\text{B}},0}}^{{{\text{C}},1}} \frac{{k_{c} }}{{\left( {\upsilon_{{N_{2} \left( {{\text{C}},1} \right)}}^{R} + [R]k_{{N_{2} \left( {{\text{C}},1} \right)}}^{{Q_{R} }} + [N_{2} ]k_{{N_{2} \left( {{\text{C}},1} \right)}}^{{Q_{{N_{2} }} }} } \right)}}$

The N₂(A) density can then be deduced from the absolute [N] concentration and the measured $\frac{{I_{580} }}{{I_{316} }}$ band intensity ratio, following:

$$a_{N + N} \frac{{I_{580} }}{{I_{316} }} = k_{3} \left( {\frac{[N]}{{[N_{2} (A)]}}} \right)^{2}$$

(3)

where $k_{3} = \frac{{k_{1} }}{{k_{2} }}$ = 2.5(± 0.5) × 10⁻⁷ in Ar/> 10%N₂ and He/> 10%N₂ and $k_{3}$ = 4 × 10⁻⁷ in Ar/(2–5%)N₂, with $\frac{{c_{580} }}{{c_{316} }}$ = 7 [8,9,10].

Assuming now that in the early afterglow, N₂(B,v = 11) levels can also be produced by reactions (b), it comes:

$$\frac{{a_{N + N} }}{{1 - a_{N + N} }} = k_{4} \frac{{[N]^{2} }}{{\left( {[N_{2} (A)][N_{2} (X,v > 13)]} \right)}}$$

(4)

where [N₂(X,v > 13)] is the sum of the densities of the N₂(X,v) vibrational levels conducing to the exothermicity of reaction (b), happening for v > 13. It is calculated $k_{4} = \frac{{k_{1} }}{{k_{b} }}$ = 7 × 10⁻⁶ in Ar/> 2%N₂ and in He/> 80%N₂, slowly decreasing to $k_{4}$ = 5 × 10⁻⁶ in He/(60–80%)N₂, $k_{4}$ = 3 × 10⁻⁶ in He/(10–40%)N₂ and $k_{4}$ = 2 × 10⁻⁶ in He/(2–5%)N₂.

Considering that the N₂⁺ (1−) (0–0) band emission at 391 nm (I₃₉₁) is produced by the following processes:

$${\text{N}}_{2}^{ + } ({\text{X}}) \, + {\text{ N}}_{2} \left( {{\text{X}},{\text{v}} > 13} \right) \, \to {\text{ N}}_{2}^{ + } \left( {{\text{B}},{\text{v}}^{{\prime }} = 0} \right) \, + {\text{ N}}_{2} ({\text{X}})$$

(d)

$${\text{N}}_{2}^{ + } ({\text{B}},{\text{v}}^{{\prime }} = 0) \, \to {\text{ N}}_{2}^{ + } ({\text{X}},{\text{v}}^{{\prime \prime }} = 0) \, + {\text{ h}}\upnu \, (1 - ,\;391\;{\text{nm}}),$$

(d′)

the N₂⁺ ion concentration can be deduced from the $\frac{{I_{391} }}{{I_{316} }}$ band intensity ratio, following the equation:

$$\frac{{I_{391} }}{{I_{316} }} = \frac{{k_{6} }}{{k_{2} }}\frac{{\left[ {N_{2}^{ + } } \right]\left[ {N_{2} (X,v > 13)} \right]}}{{\left[ {N_{2} \left( A \right)} \right]^{2} }} = k_{5} \frac{{\left[ {N_{2}^{ + } } \right]\left[ {N_{2} (X,v > 13)} \right]}}{{\left[ {N_{2} \left( A \right)} \right]^{2} }}$$

(5)

with $k_{6} = c_{391} \frac{hc}{391}A_{{{\text{X}},0}}^{{{\text{B}},0}} \frac{{k_{d} }}{{\left( {\upsilon_{{N_{2}^{ + } \left( {{\text{B}},0} \right)}}^{R} + [R]k_{{N_{2}^{ + } \left( {{\text{B}},0} \right)}}^{{Q_{R} }} + [N_{2} ]k_{{N_{2}^{ + } \left( {{\text{B}},0} \right)}}^{{Q_{{N_{2} }} }} } \right)}}$ k₅ is found to decrease from 8.5 to 4 × 10⁻² in Ar/x%N₂ and from 12 to 4 × 10⁻² in He/x%N₂ when x increases from 20 to 100% [11].

Uncertainties in species densities estimation by this line ratio methods are 30% for the N-atoms (only caused by errors made during NO titration, see for example the interesting discussion in [22]) and at least 50% for N₂(A), where additional sources of errors are related to the determination of the $a_{N + N}$ parameter and to uncertainties in the rate coefficients of Table 1. For N₂(X,v > 13) and N₂⁺, only the order of magnitude is expected because of the uncertainty increase linked to the use of the $\frac{{a_{N + N} }}{{1 - a_{N + N} }}$ ratio.

NH Radical and H-Atom Densities

As previously stated in [8,9,10], it is estimated that the NH(A,v = 0) radiative state is produced by the exothermic reaction (e), due to the endothermicity of the 3 body recombination N + H + M → NH(A,v = 0) + M:

$${\text{N}}_{2} ({\text{X}},{\text{v}} > 13) \, + {\text{ NH }} \to {\text{ N}}_{2} + {\text{ NH}}({\text{A}},{\text{v}} = 0)$$

(e)

$${\text{NH}}\left( {{\text{A}},{\text{v}} = 0} \right) \, \to {\text{ NH}}\left( {{\text{X}},{\text{v}} = 0} \right) \, + {\text{ h}}\upnu\left( {336\;{\text{ nm}}} \right)$$

(e′)

The intensity ratio method above developed can be applied to determine the absolute NH radical density, conducing to:

$$I_{336} = k_{7} \left[ {NH} \right]\left[ {{\text{N}}_{2} \left( {{\text{X}},{\text{v}} > 13} \right)} \right]$$

(6)

with $k_{7} = c_{336} \frac{hc}{336}A_{{{\text{X}},0}}^{{{\text{A}},0}} \frac{{k_{e} }}{{\left( {\upsilon_{{NH\left( {{\text{A}},0} \right)}}^{R} + [R]k_{{NH\left( {{\text{A}},0} \right)}}^{{Q_{R} }} + [N_{2} ]k_{{NH\left( {{\text{A}},0} \right)}}^{{Q_{{N_{2} }} }} } \right)}}$

$$a_{N + N} \frac{{I_{580} }}{{I_{336} }} = \frac{{k_{1} }}{{k_{7} }}\frac{{[N]^{2} }}{{\left( {[NH][N_{2} (X,v > 13)]} \right)}} = k_{8} \frac{{[N]^{2} }}{{\left( {[NH][N_{2} (X,v > 13)]} \right)}} .$$

(7)

Using $\frac{{c_{580} }}{{c_{336} }}$ = 5 [8,9,10], $A_{{{\text{X}},0}}^{{{\text{A}},0}} = \upsilon_{{NH\left( {{\text{A}},0} \right)}}^{R}$ and $k_{e}$ = 5 × 10⁻¹¹ cm³ s⁻¹ (chosen to be equal to the rates of the N₂(X,v > 13) + N₂(A) → N₂ + N₂(B,11) and N₂(X,v > 13) + N₂⁺ → N₂ + N₂⁺(B) exothermic reactions [13, 18]), it is calculated $k_{8 }$ = 1.3 × 10⁻⁷ in pure N₂ and in Ar/> 40%N₂, $k_{8}$ = 3.0 × 10⁻⁷ in Ar/< 40%N₂ and $k_{8}$ = 2.0 (± 0.5) × 10⁻⁷ in He/> 2%N₂.

The H-atom and NH radical densities are related by the following kinetics:

$${\text{N }} + {\text{ H }} + {\text{ M }} \to {\text{ NH }} + {\text{ M}}$$

(f)

$${\text{N }} + {\text{ NH }} \to {\text{ H }} + {\text{ N}}_{2}$$

(g)

Using the pseudo-stationary approximation and considering the k_f rate coefficient independent of the third body ($k_{f}^{{N_{2} }} = k_{f}^{Ar} = k_{f}^{He} )$, it comes at 8 Torr:

$$[NH] = \frac{{k_{f} }}{{k_{g} }}[H][M] = 2.6 10^{ - 4} [H]$$

(8)

Figure 2 reproduces spectra measured at z = 3 cm in the Ar/2%(N₂–5%H₂) mixture at 8 Torr, 1 slpm and 150 W.

Fig. 2

a Sequence 300–400 nm, b sequence 560–620 nm of the N₂ 1+ system observed in the Ar/2%(N₂–5%H₂) afterglow at z = 3 cm. (8 Torr, 1 slpm, 150 W, 1 ms)

In Fig. 2a, the NH (0–0) and (1–1) bands at respectively 336 and 337 nm are the most intense emissions. It is also observed the N₂ 2+ (1–0) band at 316 nm, the NO β (0–8) band at 320 nm and the N₂⁺ (0–0) band at 391 nm, close to the CN violet bands between 385 and 388 nm. For data treatment, the NH (1–1) band at 337 nm has been discarded because of possible mixing with the N₂ 2+ (0–0) band at 337 nm. The I₃₃₆ intensity is thus measured from the half intensity of the I₃₃₆ and I₃₃₇ band junction.

As previously shown by considering the I₁₁/I₉ ratio, the a_N+N coefficient can be deduced from the vibrational distribution observed in Fig. 2b [11]. With the experimental conditions of Fig. 2, it is found the value a_N+N = 0.35 and after calibration by NO titration, it is deduced [N] = 1.0 × 10¹⁵ cm⁻³.

In previous studies [11], the NH and H densities were obtained in N₂/2.5%H₂, Ar/50%(N₂–5%H₂) and He/80%(N₂–5%H₂) gas mixtures. For N₂/2.5%H₂ and Ar/50%(N₂–5%H₂), it was found a [H]/2[H₂] dissociation rate of about 0.3%. In the He/80%(N₂–5%H₂) mixture, the hydrogen dissociation rate was found to be higher (1.0%) and it was obtained a [H]/[N] ratio of about 30%.

According to Eq. (7), the accuracy of the determination of the [NH] density largely depends on the accuracy of the [N₂(X,v > 13)] density (limited to the order of magnitude) and on the choice of the k_e rate coefficient for reaction (e), chosen to be similar to the one of the N₂(X,v > 13) + N₂(A) → N₂ + N₂(B,11) and N₂(X,v > 13) + N₂⁺ → N₂ + N₂⁺(B) exothermic reactions [11].

As the H density is related to the NH density through Eq. (8) and to the k_f rate coefficients of the 3-body recombination N + H + M → NH + M, it is clear that the presented absolute densities of NH radicals and H atoms are highly speculative. Nevertheless, their relative variations with R(He or Ar) and with the dilution x of the (N₂–5%H₂) mixture in the rare gas remain significant.

Table 2 compares the a_N+N coefficients and the absolute densities obtained at z = 3 cm using the developed line ratio methods in the He/x(N₂–5%H₂) and the Ar/x(N₂–5%H₂) mixtures, for x varying between 2 and 10% (8 Torr, 1 slpm, 150 Watt).

Table 2 a_N+N coefficients, active species densities and dissociation rates determined at z = 3 cm in the early afterglows of R/x%(N₂–5%H₂) gas mixtures for x = 2–10% (8 Torr, 1 slpm and 150 W)

With He and for x < 20%, the accuracy of the abacus given in Fig. 3a [7] and showing the variation of the I₁₁/I₉ ratio is too low and it has been chosen to use instead the I₁₁/I₁₀ ratio, given in Fig. 3b.

Fig. 3

Variation of the I₁₁/I₉ (a) and I₁₁/I₁₀ (b) band intensity ratios in the mixed region between the pink afterglow (a_N+N = 0) and the full late afterglow (a_N+N = 1) for the He/x(N₂–5%H₂) mixtures, with x = 2–100% (Color figure online)

At high dilutions (2% and 5%), N-atom, N₂(A), N₂⁺ and N₂(X,v > 13) densities are higher in Ar/x(N₂–5%H₂) mixtures than in He/x(N₂–5%H₂) mixtures. At lower dilutions (20–50%), densities of N₂⁺ and N₂(X,v > 13) are also higher in the Ar mixture than in the He mixture.

Whatever the buffer gas (R = Ar or He), nitrogen and hydrogen dissociation rates are the highest in highly diluted mixtures (x = 2%). The nitrogen dissociation rate, which is usually less than 1% in pure N₂ afterglows [8,9,10], increases by one order of magnitude with high dilution in Ar, to reach 10% in the Ar/2%(N₂–5%H₂) mixture.

The H-atom concentration is minimum in medium dilution mixtures (5–10%), while the N-atom concentration is less dilution dependent. As produced by the 3-body reaction (f), the NH density also shows a minimum for (5–10%) mixtures.

Destruction of H-Atoms on the Quartz Tube Wall

From values reported in Table 3 for the Ar/2%(N₂–5%H₂) afterglow, it is deduced that the N–atom density remains practically constant along the dia.18 mm quartz afterglow tube between 3 and 35 cm (10⁻³ to 10⁻² s). At the contrary, the H-atom density shows a sensitive decrease, which is the result of a significant destruction on the tube wall.

Table 3 a_N+N factors (from the I_11/9 band ratio) and active species densities determined at z = 3, 25 and 35 cm in the afterglows of an Ar/2%(N₂–5%H₂) gas mixture (8 Torr, 1 slpm and 150 W)

Assuming that at 8 Torr, H-atoms losses in the afterglow are only due to wall recombination through H + H + wall → H₂ + wall and H + N + wall → NH + wall processes, the global $\gamma_{H}$ destruction probability on the quartz tube walls can be calculated, as demonstrated in [2] for the N atoms. It is written:

$$[H]_{z} = [H]_{z = 0} exp\left[ { - \frac{{\nu_{H} z}}{v}} \right]$$

(8)

with $\nu_{H} = \gamma_{H} \frac{\left\langle v \right\rangle }{2r}$, where $\left\langle v \right\rangle$ is the thermal gas velocity ($\left\langle v \right\rangle$ = 5 × 10⁴ cm s⁻¹ at 300 K) and r is the tube radius

z = 0 being taken at the beginning of the afterglow in the dia.18 mm tube (Fig. 1), it is obtained $\gamma_{H}^{Ar}$ = 4 × 10⁻³

As reported in Table 4 for 2% ≤ x ≤ 10% in He/x(N₂–5%H₂) mixtures, the N-atom density also shows a nearly constant value between the early and the late afterglow.

Table 4 a_N+N factors (from the I_11/10 band ratio) and active species densities determined at z = 3, 20 and 40 cm in the afterglows of He/x%(N₂–5%H₂) gas mixtures with x = 2, 5 and 10% (8 Torr, 1 slpm and 150 W)

[N₂(A)] and [N₂⁺] densities decrease sharply between z = 3 cm and z = 40 cm while the [N₂(X,v > 13)] density increases. The high concentrations of [N₂(X,v > 13)] obtained for z ≥ 20 cm are related to low densities of [N₂(A)], resulting of dominant loss frequencies of N₂(X,v > 13) and N₂(A) densities produced by reaction (b) where k_b = 4 × 10⁻¹¹ cm³ s⁻¹ [12].

Using Eq. (8), a global destruction probability of $\gamma_{H}^{He}$ = 2.5(± 1.0) × 10⁻⁴ is obtained in He/x(N₂–5%H₂) mixtures. It has been observed that $\gamma_{H}^{He}$ kept a nearly constant value for x = 2–90%. The obtained order of magnitude is in good agreement with values $\gamma_{H}^{{H_{2} }}$ = 7 × 10⁻⁴ and $\gamma_{H}^{{N_{2} /H_{2} }}$ = 4 × 10⁻⁴ previously reported in H₂ and a N₂–90%H₂ gas mixture [24]. It appears that the $\gamma_{H}^{Ar}$ value in the Ar/2%(N₂–5%H₂) gas mixtures is higher by about one order of magnitude. In comparison, the $\gamma_{N}^{{N_{2} }}$ value of the N-atoms destruction probability on quartz walls in pure N₂ is in the range 10⁻⁵–10⁻⁴ [2, 23], decreased by a factor 2 when 1% H₂ is introduced into N₂ [24].

The $\gamma_{N}^{{N_{2} }}$ value is thus about one order of magnitude lower than $\gamma_{H}^{He}$ and two orders of magnitude lower than $\gamma_{H}^{Ar}$.

Conclusion

Early afterglows of R/x%(N₂–5%H₂) (R = Ar or He) gas mixtures have been studied to obtain the absolute densities of N-atoms, N₂(A) and N₂(X,v > 13) metastable molecules by line intensity ratio methods, after calibration of the N-atom density by NO titration. The line intensity ratio method also allowed estimating the density of NH radicals and H-atoms.

For this evaluation, a rate coefficient of 5 × 10⁻¹¹ cm³ s⁻¹ has been considered for the reaction N₂(X,v > 13) + NH → N₂ + NH(A,v = 0) and a rate coefficient of 5 × 10⁻³² cm⁶ s⁻¹ for the N + H + M → NH + M 3-body recombination.

Such values conduce to high dissociation rates of H₂ and N₂ in R/x%(N₂–5%H₂) early afterglows with R = Ar or He ([H]/2[H₂] = 15% with Argon and 6% with He) for x < 5% at 8 Torr, 1 slpm, 150 W.

Using H-atoms density profiles along the afterglow quartz tubes, the global H-atom destruction probability $\gamma_{H}^{R}$ was calculated in the different R/x%(N₂–5%H₂) gas mixtures.

It is found a value $\gamma_{H}^{He}$ = 2.5(± 1.0) × 10⁻⁴ in the He/(2–90)%(N₂–5%H₂) mixtures, one order of magnitude lower than the $\gamma_{H}^{Ar}$ = 4 × 10⁻³ value obtained in the Ar/2%(N₂–5%H₂) mixture. The $\gamma_{H}^{He}$ probability is in good agreement with data published by Gordiets et al. and is higher by one order of magnitude when compared with the $\gamma_{N}^{{N_{2} }}$ recombination probability.

Appearing to be a rich source of NH radicals and H atoms, accompanying dominant N-atoms, highly diluted R/2%(N₂–5%H₂) (R = Ar or He) early afterglows are of interest for surface treatments, as already observed for selective surface nitriding of TiO₂ films [3, 4] in pure N₂ afterglow.

In future works, the N and H atoms absolute densities will be determined by TALIF to compare with the present results obtained by NO titration and the line-ratio intensity method.