Determination of the Rate Constants for the NH2 (X2B1)+ NH2 (X2B1

Jan 26, 2012 - Abstract Image. The recombination rate constants for the reactions NH2(X2B1) + ... k 0 CH4 = 9.4 × 10–29 and F cent CH4 = 0.61; C2H6...
0 downloads 0 Views 6MB Size
Article pubs.acs.org/JPCA

Determination of the Rate Constants for the NH2(X2B1) + NH2(X2B1) and NH2(X2B1) + H Recombination Reactions with Collision Partners CH4, C2H6, CO2, CF4, and SF6 at Low Pressures and 296 K. Part 2. Gokhan Altinay and R. Glen Macdonald* Chemical Sciences and Engineering Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439-4381, United States

ABSTRACT: The recombination rate constants for the reactions NH2(X2B1) + NH2(X2B1) + M → N2H4 + M and NH2(X2B1) + H + M → NH3 + M, where M was CH4, C2H6, CO2, CF4, or SF6, were measured in the same experiment over presseure ranges of 1−20 and 7−20 Torr, respectively, at 296 ± 2 K. The NH2 radical was produced by the 193 nm laser photolysis of NH3. Both NH2 and NH3 were monitored simultaneously following the photolysis laser pulse. High-resolution time-resolved absorption spectroscopy was used to monitor the temporal dependence of both species: NH2 on the 1221 ← 1331 rotational transition of the (0,7,0)A2A1 ← (0,0,0)X2B1 electronic transition near 675 nm and NH3 in the IR on either of the inversion doublets of the q Q3(3) rotational transition of the ν1 fundamental near 2999 nm. The NH2 self-recombination clearly exhibited falloff behavior for the third-body collision partners used in this work. The pressure dependences of the NH2 self-recombination rate constants were fit using Troe’s parametrization scheme, kinf, k0, and Fcent, with kinf = 7.9 × 10−11 cm3 molecule−1 s−1, the theoretical value calculated by Klippenstein et al. (J. Phys. Chem. A 113, 113, 10241). The individual Troe parameters were CH4, k0CH4 = 9.4 × CH4 C2H6 CO2 10−29 and Fcent = 0.61; C2H6, k0C2H6 = 1.5 × 10−28 and Fcent = 0.80; CO2, k0CO2 = 8.6 × 10−29 and Fcent = 0.66; CF4, k0CF4 = 1.1 × −28 −28 6 −2 −1 CF4 SF6 SF6 10 and Fcent = 0.55; and SF6, k0 = 1.9 × 10 and Fcent = 0.52, where the units of k0 are cm molecule s . The NH2 + H + M reaction rate constant was assumed to be in the three-body pressure regime, and the association rate constants were CH4, (6.0 ± 1.8) × 10−30; C2H6, (1.1 ± 0.41) × 10−29; CO2, (6.5 ± 1.8) × 10−30; CF4, (8.3 ± 1.7) × 10−30; and SF6, (1.4 ± 0.30) × 10−29, with units cm6 molecule−1 s,−1 and the systematic and experimental errors are given at the 2σ confidence level. species needs to be known.2−9 However, these conditions are in general not easily achieved. Alternatively, the study of radical selfreactions requires only the determination of the concentration of a single transient species, and these reactions form a small subset of radical−radical reactions that have received the most scrutiny. The self-radical reactions of the isoelectronic radicals CH3(X2A″2), OH(X2Π), and NH2(X2B1) have been extensively studied both experimentally10−14 and theoretically.15−23 The self-radical reactions involving CH3 and OH radicals are taken to be prototypical of radical−radical reactions. All three radicals play an important role in combustion chemistry.24

I. INTRODUCTION Radical−radical reactions are a unique class of chemical reactions that provide challenges for both their experimental measurement and the theoretical description of their behavior.1 For the experimentalist, the challenge is to measure the temporal concentration profiles of two transient species to determine a rate constant. If product branching ratios are desired, other transients or stable species concentrations must also be determined. Almost all bimolecular atom/radical plus molecule rate constants are determined under pseudo-first-order conditions in which only the temporal dependence of the transient species needs to be known, not its absolute concentration. It is possible to arrange these conditions in the study of radical−radical reactions in which one transient species concentration is much larger than the other and only the concentration of the excess © 2012 American Chemical Society

Received: December 20, 2011 Revised: January 24, 2012 Published: January 26, 2012 2161

dx.doi.org/10.1021/jp212280q | J. Phys. Chem. A 2012, 116, 2161−2176

The Journal of Physical Chemistry A

Article

Figure 1. Typical NH2 and NH3 temporal concentration profiles obtained after the 193 nm photolysis laser pulse with CH4 as a collision partner is shown. In panel a, the experimental NH2 profile is shown by the solid red line, the open black circles are the optimum fit to the NH2 profile by varying k1b and k2b (see text), and the solid green circles are the model prediction for the H atom concentration profile. The variation in the H atom concentration as a function of kdiff(H) is shown by the blue dashed line for kdiff(H) increased by 50%, and by the black dashed−dotted line for kdiff(H) decreased by 50%. The corresponding changes in k2b were ∓15%. Panel b shows the simultaneously measured NH3 profile. The solid red line is the experimental NH3 profile, and the open black circles are the optimum fit to the NH3 profile varying k1b and k2b. The blue dashed line shows only the removal by kdiff(NH3). The reaction path analysis showed that the fraction of NH2 radicals lost by reactions 1b and 2b was 0.76 and 0.041, respectively, and the fraction of NH3 produced by reaction 2b was 0.089, with diffusion and flow accounting for the remainder. The conditions of the experiment were PCH4 = 10.33 and PNH3 = 0.0095 Torr at 295 K. (c) The same as part a, except the pressures were PCH4 = 17.41 and PNH3 = 0.012 Torr. The variation in kdiff(H) of ±50% resulted in a variation in k2b of ∓11%. The reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.79 and 0.11, respectively. (d) Same as panel b, except the NH3 profile is the simultaneously recorded profile for the NH2 profile in panel c. The fraction of NH3 produced by reaction 2b was 0.23 with diffusion and flow accounting for the remainder.

undergone experimental2,10,25 and theoretical investigation.26−29 They have the advantage that the PES of a radical + H atom reduces the dimensionality of the system significantly; thus, increasing the accuracy of the PES describing the radical−atom interaction and making dynamical approximations more realistic.30 At low temperatures, only the recombination channel leading to singlet products, CH4, H2O, and NH3, are energetically accessible. Again, CH3 and OH radicals are dominant radical species in combustion, and their corresponding interactions with H atoms have received more attention than that with the NH2 radical. Indeed, there has been only a

All three of these self-radical interactions involve multiple spin manifolds. The interaction of two radicals in doublet states produces singlet and triplet potential energy surfaces (PESs). At temperatures near 300 K, self-recombination on the singlet PES leads to the singlet products C2H6, H2O2, and N2H4, respectively, and on the triplet PES, the products, H2O + O(3P) and NH3 + NH(X3Σ−) are energetically accessible for the OH and NH2 radicals, respectively. The CH3, NH2, and OH + H atom reactions are also a class of atom−radical reactions that are important in combustion24 and other chemical environments. These reactions have also 2162

dx.doi.org/10.1021/jp212280q | J. Phys. Chem. A 2012, 116, 2161−2176

The Journal of Physical Chemistry A

Article

Figure 2. Same as Figure 1, except C2H6 is the collision partner. (a) The reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.79 and 0.11, respectively. The variation in kdiff(H) of ±50% resulted in a variation in k2b of ∓16%. The conditions of the experiment were PC2H6 = 8.19 and PNH3 = 0.021 Torr at 296 K. (b) The fraction of NH3 produced in reaction 2b was 0.22, with diffusion and flow accounting for the remainder. (c) The reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.76 and 0.17, respectively. A variation in kdiff(H) of ±50% resulted in a variation in k2b of ∓9%. The conditions of the experiment were PC2H6 = 17.09 and PNH3 = 0.015 Torr at 296 K. (d) The fraction of NH3 produced in reaction 2b was 0.35, with diffusion and flow accounting for the remainder.

single low-temperature measurement of the NH2 + H → NH3 recombination reaction.25 The role of NH2 in combustion chemistry has been elucidated and summarized by Miller and Bowman31 and stems from its participation in NOX chemistry; however, the NH2 radical is also important in the thermal decomposition of ammonium32−34 and hydrazine.35 With the increased need to reduce the accumulation of greenhouse gases in the atmosphere, biomass combustion will play an important role in the mix of future energy sources. Thus, the gas phase evolution of NH3 from biomass combustion will increase the importance of NH3 and NH2 chemistry to the whole combustion process.36 A number of recent studies have been directed at the interaction of NH3 and NH2 chemistry with simple hydrocarbon fuels.37 The present work is a continuation of previous work38 from this laboratory on the radical−radical chemistry of NH2 with transient species and the study of the self-recombination of the NH2 radical in the presence of a variety of third bodies.13,14 As well, the high signal-to-noise ratio of the recorded concen-

tration profiles allowed for the determination of the NH2 + H recombination rate constant with six collision partners (including part 1). Although there have been numerous studies of the corresponding isoelectronic radical (CH3 and OH) recombination reactions, as noted above, virtually all previous investigations have been limited to a few collision partners; mainly, He, Ar, and N2. This makes it difficult to understand the dynamics of the recombination process with such a limited data set of structurally similar collision partners. The collision partners chosen in this workCH4, C2H6, CO2, CF4, and SF6were taken as representative of a variety of molecular shapes, internal state density, and molecular weight. Furthermore, the pressure range was sufficient that the limiting low-pressure rate constant, k0, could be determined by a Troe fit to the bimolecular rate constant pressure dependence. These measurements provide a consistent data set to compare with theoretical predictions so that the underlying factors governing energy transfer efficiencies in unimolecular reactions could be examined. 2163

dx.doi.org/10.1021/jp212280q | J. Phys. Chem. A 2012, 116, 2161−2176

The Journal of Physical Chemistry A

Article

Figure 3. Same as Figure 1, except CO2 is the collision partner. (a) The reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.84 and 0.095, respectively. A variation in kdiff(H) of ±50% produced a variation in k2b of ∓12%. The condition of the experiment was PCO2 = 11.48 and PNH3 = 0.017 Torr at 295 K. (b) The fraction of NH3 produced in reaction 2b was 0.27, with diffusion and flow accounting for the remainder. (c) A reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.81 and 0.13, respectively. The variation in kdiff(H) of ±50% resulted in a variation in k2b of ∓10%. The conditions of the experiment were PCO2 = 18.28 and PNH3 = 0.026 Torr at 295 K. (d) The fraction of NH3 produced in reaction 2b was 0.36, with diffusion and flow accounting for the remainder.

NH2OH and reported on the theoretical estimates of rate constants for a number of important chemical reactions contributing to the secondary chemistry in the NH2OH system over a temperature range of 200−2000 K. These theoretical calculations yielded Troe parameters39,40 kinf, k0, and Fcent as a function of temperature for reaction 1b with N2 as a collision partner. The stationary points on both the singlet and triplet PESs for the NH2−NH2 interaction were examined at a high level of electronic structure theory, CCSDT(T)/CBS//CCSD(T)/aug-cc-pvdz, coupled with CASPT2(4e,4o) calculations to refine the barrier height on the triplet PES. These calculations clearly establish that the barrier heights leading to N2H2(isomers) + H2 are too high for these channels to contribute near 300 K, greatly simplifying the interpretation of the experimental measurements. This work also provided theoretical estimates for the

The low-pressure recombination rate constants were measured for reactions 1b and 2b, where M is the collision partner for the reaction, CH4, C2H6, CO2, CF4, or SF6, and the reaction numbers correspond to the numbering in Section III. In our initial study,14 part 1, we reported on the measurement of the low-pressure dependence of the rate constants for reaction 1b with He, Ne, Ar, and N2 as collision partners and the determination of the association rate constant for reaction 2b with N2 as a third body. There have been previous experimental studies of reaction 1b near 300 K, and we have recently reported13 on and discussed previous measurements of k1b with Ar, N2, and CF4. The values of k1b from the new study were significantly larger than the older results. Klippenstein et al.22 conducted a detailed experimental and theoretical study of the thermal decomposition of 2164

dx.doi.org/10.1021/jp212280q | J. Phys. Chem. A 2012, 116, 2161−2176

The Journal of Physical Chemistry A

Article

Figure 4. Same as Figure 1, except CF4 is the collision partner. (a) The reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.81 and 0.11, respectively. The variation in kdiff(H) of ±50% resulted in a variation in k2b of ∓11%. The conditions of the experiment were PCF4 = 7.02 and PNH3 = 0.012 Torr at 295 K. (b) The fraction of NH3 produced in reaction 2b was 0.32, with diffusion and flow accounting for the remainder. (c) The reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.78 and 0.19, respectively. The variation in kdiff(H) of ±50% resulted in a variation in k2b of ∓6%. The conditions of the experiment were PCF4 = 20.82 and PNH3 = 0.018 Torr at 296 K. (d) The fraction of NH3 produced in reaction 2b was 0.64, with diffusion and flow accounting for the remainder.

abstraction channel rate constant on the triplet surface. At 296 K, the calculated value was 1.8 × 10−15 cm3 molecule−1 s−1, and it also will not significantly contribute to NH2 removal. Reaction 2b has not received the same attention as the corresponding reactions of H atoms with CH3 and OH radicals. The only previous study was by Gordon et al.25 The NH2 radical was created by pulse radiolysis of NH3 and detected by white light absorption spectroscopy on the partially resolved (0,8,0) A2A1 ← (0,0,0) X2B1 transition. The reverse reaction has been studied32 at high temperature, but not extensively. The authors are unaware of any theoretical work on reaction 2b although there have been studies on reaction 2a.22,28 A recent estimate of the value of the abstraction channel rate constant on the triplet PES was measured by Bahng and Macdonald38 to be 8 × 10−15 cm3 molecule−1 s−1 at 293 K. This measurement is lower than an extrapolation of a theoretical high-temperature calculation by Linder et al.28

As in part 1, the NH2 radical was produced by 193 nm laser photolysis of NH3 dilute in the collision partner gas. The temporal concentrations of both NH2 and NH3 were monitored by high-resolution laser absorption spectroscopy. The NH2 radial was detected by the 1231 ← 1331 rotational transition of the (0,7,0) A2A1 ← (0,0,0) X2B1 band near 675 nm, and NH3 by either inversion doublet of the qQ3(3) rotational transition of the υ1 fundamental near 3000 nm. Both species were detected simultaneously. The absorption signal for the NH2 radical was directly related to the loss of NH3 by photolysis because the quantum yield is almost 1.38

II. EXPERIMENTAL SECTION The experimental apparatus has been described13,14,38 and will not be discussed in detail here. Briefly, an inner Teflon box was removed to increase the immediate chamber volume accessible 2165

dx.doi.org/10.1021/jp212280q | J. Phys. Chem. A 2012, 116, 2161−2176

The Journal of Physical Chemistry A

Article

Figure 5. Same as Figure 1, except SF6 is the collision partner. (a) The reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.75 and 0.21, respectively. The variation in kdiff(H) of ±50% resulted in a variation in k2b of ∓11%. The conditions of the experiment were PSF6 = 11.5 and PNH3 = 0.015 Torr at 295 K. (b) The fraction of NH3 produced in reaction 2b was 0.48, with diffusion and flow accounting for the remainder. (c) The reaction path analysis showed that the fraction of NH2 radicals removed by reactions 1b and 2b was 0.68 and 0.29, respectively. The variation in kdiff(H) of ±50% resulted in a variation in k2b of ∓4%. The conditions of the experiment were PSF6 = 22.5 and PNH3 = 0.016 Torr at 296 K. (d) The fraction of NH3 produced in reaction 2b was 0.70, with diffusion and flow accounting for the remainder.

capture absorption signals from both laser systems. The absorption signals from each probe laser were monitored by dual detectors to record the transient signal and the initial laser intensity. Common-mode noise was suppressed by direct subtraction using software. The tuning of each laser to the line center absorption of the monitoring transition was enhanced over earlier work from this laboratory using a boxcar average to record the maximum absorption signal (only a 20 μs window). In the present work and part 1, complete transient absorption signals were integrated using a 200 kHz 16 bit A/D 8 channel signal acquisition board, greatly increasing the signal-to-noise ratio of the tuning signal. The output of this tuning hardware was averaged for a set number of laser pulses and monitored continuously.

to reaction products and provide a possible sink for reaction products on the stainless steel walls. The gases used, with supplier and stated purity, were CH4, Linde, 99.99%; C2H6, Linde, 99.95%; CO2, Air Gas, 99.999%; CF4, Linde, 99.99%; SF6, Air Gas, 99.99%; and NH3, AGA, 99.99%. The absorption path length was increased by using White cell optics to multipass the probe laser radiation through the photolysis zone. The photolysis laser was a Lambda Physik Compex 205. The NH2 radical was monitored using an external cavity diode laser from Sacher Lasertechnik. The NH3 molecule was monitored by a Linos model 4500 OPO system modified to operate in a dual cavity configuration. The data acquisition hardware and software were improved from previous experiments.14 High-resolution 14 and 16−22 flexible bit A/D dual-channel transient recorders were used to 2166

dx.doi.org/10.1021/jp212280q | J. Phys. Chem. A 2012, 116, 2161−2176

The Journal of Physical Chemistry A

Article

Table 1. Summary of Experimental Conditions and Measurements of k1b and k2b in CH4 at 296 ± 2 K partial pressure (Torr) PCH4 1.22 1.32 1.38 1.43 1.51 1.61 1.87 2.05 2.18 3.18 3.57 3.84 5.68 6.38 6.85 8.57 9.66 10.33 11.3 12.57 13.42 14.38 17.41 a

PNH3 0.0153 0.017 0.0193 0.019 0.0193 0.017 0.0152 0.017 0.017 0.0106 0.0138 0.012 0.0084 0.011 0.0095 0.0098 0.006 0.0072 0.0116 0.009 0.0066 0.0072 0.0091

range [NH2]0a,b (× 10−13)

prbd

2.2−0.24 2.4−0.24 2.5−0.25 2.5−0.28 2.6−0.26 2.4−0.27 2.2−0.26 2.5−0.2 2.4−0.27 1.81−0.21 2.1−0.24 2−0.26 1.34−0.12 1.4−0.17 1.4−0.18 2−0.18 1.3−0.11 1.1−0.11 1.8−0.19 1.8−0.14 1.26−0.14 1.2−0.09 1.56−0.18

1.14 1.13 1.22 1.22 1.16 1.1 1.08 1.08 1.1 0.94 0.85 0.94 0.82 0.79 0.84 0.82 0.75 0.76 0.77 0.76 0.76 0.74 0.73

k1bc (× 1011)

k2bc (× 1012)

d

0.37 (±0.040) 0.442 (±0.028) 0.437 (±0.050) 0.338 (±0.057) 0.377 (±0.041) 0.423 (±0.026) 0.512 (±0.045) 0.496 (±0.058) 0.503 (±0.065) 0.704 (±0.070) 0.72 (±0.11) 0.743 (±0.12) 0.864 (±0.19) 1.04 (±0.089) 0.97 (±0.57) 1.23 (±0.14) 1.50 (±0.095) 1.48 (±0.089) 1.54 (±0.051) 1.68 (±0.13) 1.86 (±0.12) 1.65 (±0.12) 2.12 (±0.092)

2.48 (±0.31) 1.41 1.14 (±0.23) 2.31 (±0.19) 3.37 (±0.88) 3.09 (±1.19) 3.39 (±0.22)

Multiple rate constant measurements were made under the same flow conditions. bUnits = molecules cm.−3. cUnits = cm3 molecule−1 s.−1 Uncertainty is ±1σ in the scatter of these measurements.

d

Table 2. Summary of Experimental Conditions and Measurements of k1b and k1b in C2H6 at 296 ± 2 K partial pressure (Torr) PC2H6 0.63 0.65 0.73 1.09 1.1 1.27 1.85 2.13 2.32 3.99 4.2 4.7 7.09 7.33 8.19 10.56 10.65 12.14 14.69 15.02 17.09 a

PNH3 0.012 0.039 0.015 0.016 0.014 0.014 0.015 0.012 0.012 0.0090 0.0090 0.0077 0.017 0.0084 0.017 0.0072 0.0078 0.010 0.0090 0.0098 0.0098

range [NH2]0a,b (× 10−13)

prbd

1.6−0.18 2.1−0.24 2.2−0.23 2.4−0.27 1.8−0.2 2.3−0.25 2.4−0.25 1.9−0.21 1.7−0.19 2.0−0.22 1.5−0.18 1.6−0.17 1.8−0.2 1.1−0.12 2.1−0.24 1.4−0.16 1.5−0.17 1.5−0.14 1.8−0.21 1.4−0.16 1.2−0.15

1.2 1.27 1.3 1.17 1.1 1.16 1.1 1.01 0.98 0.87 0.85 0.85 0.8 0.75 0.79 0.73 0.71 0.72 0.69 0.65 0.65

k1bc (× 1011) 0.314 (±0.0075) 0.569 (±0.072) 0.524 (±0.087) 0.600 (±0.057) 0.576 (±0.057) 0.601 (±0.050) 0.798 (±0.037) 0.844 (±0.064) 0.803 (±0.031) 1.29 (±0.078) 1.35 (±0.058) 1.52 (±0.021) 1.8 (±0.13) 1.91 (±0.059) 1.96 (±0.11) 2.57 (±0.13) 2.65 (±0.13) 2.64 (±0.20) 3.01 (±0.16) 2.95 (±0.12) 3.53 (±0.19)

k2bc (× 1012) d

1.13 2.04 1.79 4.36 4.47 (±0.52) 4.47 3.71 (±0.77) 6.06 (±2.8) 8.11 (±3.1)

Multiple rate constant measurements were made under the same flow conditions. bUnits = molecules cm−3. cUnits cm3 molecule−1 s−1. Uncertainty is ±1 σ in the scatter of the measurements.

d

squares of the residuals (χ2) was found.13 In the present work, the NH2 profile defined the determination of k1b; the NH3 profile, the determination of k2b. The optimum rate constants for both reactions 1b and 2b were determined by successive iterations between fitting the two profiles until convergence. The

III. RESULTS AND DISCUSSION A. Reaction Mechanism. Rate constants were determined by comparing the measured temporal concentration profiles to calculated ones based on a detailed reaction mechanism. Only a single rate constant was varied until a minimum in the sum of the 2167

dx.doi.org/10.1021/jp212280q | J. Phys. Chem. A 2012, 116, 2161−2176

The Journal of Physical Chemistry A

Article

Table 3. Summary of Experimental Conditions and Measurements of k1b and k2b in CO2 at 296 ± 2 K partial pressure (Torr)

a

PCO2

PNH3

range [NH2]0a,b (× 1013)

prbd

k1bc (× 1011)

k2bc (× 1012)

0.66 0.74 1.22 1.32 2.09 2.37 4.40 5.04 7.54 8.89 11.49 13.35 15.17 18.28

0.017 0.014 0.019 0.016 0.015 0.017 0.011 0.0098 0.015 0.0084 0.011 0.012 0.011 0.014

2.4−0.25 2.1−0.23 2.7−0.3 2.4−0.24 2.4−0.18 2.3−0.26 2.5−0.24 2.1−0.24 2.4−0.22 1.4−0.12 2.2−0.25 2.4−0.28 2.2−0.22 2.3−0.23

1.79 1.48 1.45 1.26 1.10 1.13 0.88 0.84 0.78 0.65 0.66 0.63 0.59 0.54

0.12 0.169 (±0.042)d 0.269 (±0.056) 0.376 (±0.040) 0.417 (±0.053) 0.459 (±0.04) 0.868 (±0.048) 0.879 (±0.11) 1.16 (±0.13) 1.20 (±0.11) 1.62 (±0.07) 1.76 (±0.085) 1.85 (±0.033) 1.99 (±0.057)

0.325 (±0.16) 0.663 1.84 (±0.43) 2.46 (±0.34) 2.08 (±0.30) 1.67 (±0.23) 3.73 (±0.25) 4.10 (±1.1)

Multiple rate constant measurements were made under the same flow conditions. bUnits = molecules cm−3. cUnits = cm3 molecule−1 s−1. Uncertainty is ±1σ in the scatter of these measurements.

d

Table 4. Summary of Experimental Conditions and Measurements of k1b and k2b in CF4 at 296 ± 2 K partial pressure (Torr) PCF4 0.78 0.97 1.52 1.52 1.89 2.42 2.63 3.67 5.59 5.77 5.84 7.03 10.15 10.18 10.93 12.01 14.00 14.72 17.48 20.6 20.83 21.00 a

PNH3

range [NH2]0a,b (× 1013)

0.014 0.0090 0.015 0.018 0.013 0.013 0.013 0.011 0.018 0.016 0.014 0.010 0.019 0.019 0.021 0.0093 0.014 0.013 0.0090 0.015 0.013 0.013

prbd

2.4−0.27 4.1−1.1 2.7−0.28 1.6−0.18 5.1−1.4 1.3−0.17 2.9−0.32 4.3−1.16 1.3−0.14 1.7−0.19 2.5−0.27 4.1−1.1 2.3−0.25 3.0−0.33 3.0−0.4 3.2−0.86 2.5−0.26 1.6−0.2 4.5−1.3 1.8−0.22 5.3−1.4 2.2−0.24

1.17 1.13 1.05 1.10 1.07 0.99 0.98 0.93 0.896 0.88 0.85 0.86 0.77 0.79 0.73 0.76 0.74 0.73 0.73 0.67 0.69 0.66

k1bc (× 1011) 0.278 (±0.033) 0.247 (±0.056) 0.498 (±0.071) 0.413 (±0.024) 0.469 (±0.052) 0.509 (±0.040) 0.546 (±0.028) 0.842 (±0.051) 0.910 (±0.14) 0.903 (±0.082) 1.03 (±0.029) 1.26 (±0.041) 1.38 (±0.022) 1.38 (±0.0301) 1.45 (±0.073) 1.70 (±0.067) 1.66 (±0.026) 1.75 (±0.069) 2.21 (±0.066) 1.99 (±0.076) 2.37 (±0.046) 1.96 (±0.037)

k2bc (× 1012) d

1.39 (±0.45)d 2.26 (±0.099) 3.10 (±1.3) 1.63 (±0.035) 3.11 (±1.3) 2.39 (±1.6) 1.86 (±0.23) 2.71 (±0.55) 4.51 (±0.67) 4.17 (±0.47) 4.17 (±0.39) 5.21 (±0.43) 4.66 (±0.31) 5.74 (±0.53) 5.06 (±1.3)

Multiple rate constant measurements were made under the same flow conditions. bUnits molecules = cm−3. cUnits = cm3 molecule−1 s−1. Uncertainty is ±1σ in the scatter of these measurements.

d

NH2 + H → NH(X3Σ−) + H2

detailed reaction mechanism consisted of 29 reactions and 12 species and has been discussed.13,14 As given in part 1, a much simpler mechanism that accounts for 99% of the chemistry is

ΔH00r = −46 kJ mol−1

(2a)

193 nm

NH3 ⎯⎯⎯⎯⎯⎯⎯⎯→ NH2 + H NH2 + NH2 → NH(X3Σ−) + NH3 ΔH00r = −58 kJ mol−1

NH2 + H + (M) → NH3 + (M) ΔH00r = −444 kJ mol−1

(1a)

NH2 + NH2 + (M) → N2H 4 + (M) ΔH00r = −268 kJ mol−1

NH2 + NH → N2H2 + H2 (1b)

(2b)

ΔH00r = −124 kJ mol−1 (3)

2168

dx.doi.org/10.1021/jp212280q | J. Phys. Chem. A 2012, 116, 2161−2176

The Journal of Physical Chemistry A

Article

Table 5. Summary of Experimental Conditions and Measurements of k1b and k2b in SF6 at 296 ± 2 K partial pressure (Torr) PSF6 0.85 1.55 1.68 2.34 2.58 2.83 2.88 5.42 6.47 6.63 6.69 9.39 10.93 11.51 11.55 14.39 15.54 16.65 16.87 16.87 19.78 22.5 a

range [NH2]a,b (× 10−13)

PNH3 0.013 0.018 0.0023 0.012 0.011 0.0014 0.0047 0.0078 0.0088 0.002 0.0063 0.019 0.014 0.011 0.0044 0.021 0.011 0.018 0.0059 0.0071 0.026 0.010

prbd

2.1−0.21 2.3−0.48 0.57−0.073 1.3−0.14 2.0−0.23 0.33−0.042 0.96−0.10 2.3−0.23 1.8−0.27 0.63−0.073 1.7−0.27 3.0−0.36 2.5−0.32 2.6−0.28 0.89−0.12 2.7−0.29 1.8−0.29 3.2−0.4 1.3−0.17 1.32−0.23 1.7−0.17 2.0−0.23

1.01 0.95 0.78 0.83 0.89 0.73 0.79 0.81 0.77 0.73 0.75 0.74 0.74 0.72 0.68 0.69 0.68 0.69 0.68 0.67 0.63 0.65

k1bc (× 1011)

k2bc (× 1012) d

0.174 (±0.065) 0.57 (±0.069) 0.674 (±0.016) 0.0685 (±0.078) 0.800 (±0.012) 0.906 (±0.015) 1.02 (±0.057) 1.22 (±0.027) 1.55 (±0.020) 1.62 (±0.092) 1.55 (±0.021) 1.68 (±0.061) 1.97 (±0.054) 2.06(±0.043) 1.98 (±0.32) 2.01 (±0.050) 2.20 (±0.15) 2.38 (±0.12) 2.36 (±0.098) 2.38 (±0.031) 2.45 (±0.050) 2.63 (±0.12)

5.24 (±0.94) 4.28 (±0.16) 4.00 (±1.36) 3.91 (±0.98) 5.34 (±1.31) 9.27 (±2.45) 4.59 (±0.83) 8.51 (±1.1) 5.81 (±0.41) 7.44 (±2.61) 7.85 (±1.59) 8.37 (±1.48) 10.9 (±0.37)

Multiple rate constant measurements were made under the same flow conditions. bUnits = molecules cm−3. cUnits = cm3 molecule−1 s−1. Uncertainty is ±1σ in the scatter of these measurements.

d

Table 6. Summary of the Determination of bL0 for the NH3 Q3(3) ν1 Fundamental Transition

and subsequence chemistry was included in the detailed mechanism. The N2H3 radical could be produced in abstraction reactions of H or NH2 with N2H4, but these rate constants have been either measured42 or calculated theoretically,43 and are too small near 300 K to contribute. There is also an abstraction reaction between NH2 and CH4 or C2H6, but again, the rate constants are too small44 at 296 K to be significant. The importance of diffusion in the reaction scheme has been discussed.14 First-order diffusion rate constants were calculated using the diffusion-volume analysis of Fuller et al.45 This prescription provided a uniform method for calculating binary diffusion constants for species X (D12(X)). The diffusion rate constants were then determined by normalization to D12(NH3) and the measured diffusion rate constant for NH3. Unfortunately, the temporal concentration profile for diffusion of NH3 was described by two exponential terms due to the rectangular geometry, ∼1.5 × 10 cm, of the photolysis zone.13 To overcome this problem, the NH3 temporal profile was approximated by a single exponential term for a shorter period of time than the recorded profile. The NH3 removed by photolysis was replenished by diffusion and flow from the surrounding gas, resulting in a production process for NH3. Depending on the collision partner, above ∼5 Torr, reaction 2b contributes measurably to the production of NH3. Under these circumstances the determination of kdiff(NH3) was made by fitting the NH3 concentration profile after a suitable delay, usually 5 ms. As can be seen from the corresponding NH2 profile, the NH2 concentration has decreased sufficiently that reaction 2b cannot contribute to the production of NH3. In general, the reaction was recorded for a total time of 40 ms. Reaction 2b does not interfere with the determination of kdiff(NH3). B. Determination of k1b and k2b. Typical NH2 and NH3 temporal concentration profiles are shown in Figures 1−5 for

q

bL0 a collision partner pressure (Torr) 5 10 15 20 av bL0

CH4 (10−4)

C2H6 (10−4)

1.80 1.33 1.22 1.20 1.25 ± 0.14b,c

2.19 1.69 1.62 1.64 1.65 ± 0.08

CO2 (10−4) CF4 (10−4) SF6 (10−4) 2.13 2.02 2.19 2.53 1.40 ± 0.52

1.48 1.22 1.18 1.20 1.20 ± 0.04

2.44 1.69 1.48 1.42 1.52 ± 0.28

Units = cm−1 Torr−1. bThe average is from the three highest pressures. cUncertainty ±2σ in the scatter of the measurements. a

NH2 + RH → NH3 + R ΔH00r = −11.9, 28.1 kJ mol−1

(4)

X → diffusion/flow

(5) 22

where the enthalpies were taken from Klippenstein et al.; M is the third-body collision partner for recombination; R is CH3 or C2H5, respectively; and X is any species in the system. The photodissociation of NH3 at 193 nm produces NH2 with substantial excitation in electronic, vibrational, and rotational degrees of freedom.41 As well, NH3 is also produced with excess internal energy.13 The consequences of internal energy relaxation on the observed NH2 and NH3 concentration measurements have been addressed and may influence rate constant measurements at pressures less than a few Torr. Not shown in the above reaction scheme are reactions removing the NH radical and chemistry associated with the N2H3 radical. The NH radical is produced38 in the initial photolysis step, but its quantum yield has been shown to be