J. Phys. Chem. 1988, 92, 5133-5141 of two prototype structures that are illustrated by those observed in phenoxyl and p-benzosemiquinone anion radicals. One observes a progression from the phenoxyl to the semiquinone anion structure as the substituents become more capable of donating electrons to the phenyl ir system. This change is characterized by increased enhancement of the and vg, vibrations in the more semiquinone anion like structures. The resonance Raman features of non-oxy radicals can also be described in terms of these prototype structures. For example, the aniline radical cation8 is structurally a phenoxyl-like radical, while the p-phenylenediamine cationg0is a semiquinone anion like radical.
5133
The para-substituted phenoxyl radicals studied here provide model systems for discussing the resonance enhancement of the C H bendings modes, a study of some general interest. The enhancement of the vgaand ulSa vibrations, which are predominantly C H bending vibrations and have totally symmetric character under C2, symmetry, parallels that of the v8, and v , vibrations. ~ Since the latter two vibrations are frequently used to examine the structures of phenyl-related transients, observations on the related C H bending modes can be used to provide confirmation of their assignments. Registry No. 4-H3CC6H40', 3174-48-9; 4-FC6H40', 2145-21-3; 4C1C6H40', 3148-13-8; 4-BrC6H40', 63125-13-3; 4-H,COC6H40', 61 19-32-0.
(30) Hester, R. E. Adu. Infrared Raman Spectrosc. 1976, 4 , 317.
Chemical Generation of Electronically Excited N, in the H(D) 4- NF, Flame Floyd E. Hovis, Philip D. Whitefield,* Harvey V. Lilenfeld, and Gregory R. Bradburn McDonnell Douglas Research Laboratories, St. Louis, Missouri 631 66 (Received: September 8, 1987; I n Final Form: January 14, 1988)
The chemistry of the H(D) + NF2 flame has been investigated through measurements of key population profiles, production efficiencies, and rate coefficients. The absolute population profiles of N2(B), NF(a), and NF(b) and upper limits on Nz(A) concentrations generated in the H(D) + NF, system are reported. From these profiles the efficiencies of NF(a) and N2(A) and N2(B) production were determined. For NF(a) the peak concentrations, which should also represent the production efficiency, were found to be 15-30% of the input NF,. From the results of measurements by other investigatorson the branching ratio for the H + NF, reaction, we would have expected the peak [NF(a)] to be -90% of the input NF,. Relative to N2F4, the N2(B)production efficiencies were 2-4%. However, relative to peak [NF(a)] they were 20-30%. The upper limits on [N2(A)] are consistent with N,(A) production by radiative cascading from N,(B). The room-temperature rate constant for the reaction N(,D) + NF(a) F N,(B,W) was estimated to be 1 X cm3 s-I. The quenching of N,(A) by H atoms was found to have a rate constant > 5 X lo-', cm3 s-' and, as such, represents a major loss channel for N2(A).
-+
Introduction The lifetime, high electronic energy, and capability of chemical production makes N2(A3Zu+)molecules eligible for consideration as an energy storage/transfer partner for a high-energy chemical laser device. The N2(A3&+)molecule has the potential to transfer so much energy in one step [-6 times that transferred by an 02(a'A) molecule] that an extensive range of candidate lasing species can be considered for energy transfer from N2(A). The development of a laser device based on such an energy-transfer process, however, is predicated on the development of an efficient N,(A) production scheme. The reaction of H atoms with NFz radicals produces metastable NF(alA) radicals and vibrationally excited H F as primary products:1,2 H
+ NF,
-
HF
+ NF
(1)
The rate constant for reaction 1 has been measured to be (1.5 f 0.2) X lo-'' cm3 s-', with a branching ratio of 0.9 for the NF(a'A) p r o d ~ c t . ~ This - ~ reaction initiates a series of highly exothermic secondary reactions producing such species as N(,D), N(,P), N(4S), F(2P), NF(bIZ+), NF(a'A), NF(X3Z-), N2(B311,), N,(A3&+), N2(a'II,), Nz(X'Zg+), NH(c'n), "(a' A), NH(A311), and NH(X3Z-).'-3 The production of metastable N2(A38,+)in the H NFz system has been inferred from the strong Nz(B+A) first positive band emission observed under excess H reaction
+
(1) Clyne, M. A. A.; White, I. F. Chem. Phys. Lett. 1970, 6, 465. (2) Herbelin, J. M.; Cohen, N. Chem. Phys. Lett. 1973, 20, 605. (3) Cheah, C. T.; Clyne, M. A. A.; Whitefield, P. D. J . Chem. SOC., Faraday Trans. 2 1980, 76, 7 11. (4) Cheah, C. T.; Clyne, M. A. A. J . Chem. SOC.,Faraday Trans. 2 1980,
76, 1543. (5) Malins, R. J.; Setser, D. W. J . Phys. Chem. 1981, 85, 1342.
0022-3654/88/2092-5133$01.50/0
conditions3 The production of the electronically excited states of N 2 is thought to occur via the following reaction sequence:
-
+ NF(a) N(2D) + NF(a) H
N2(B)
--*
+ N(2D) N2(B,W) + F N2(A) + hv HF
(2) (3)
(4)
Radiative loss of N2(A) to the N2(X) ground state is not important, considering the metastability of N2(A). Thus, under conditions in which wall losses can be eliminated or minimized, the dominant loss channel for N2(A) will be collisional deactivation. Little is known about the efficiency of N2(A) production in the H + NF, flame. This paper presents the results of an evaluation of the H + NF, system as an efficient generator of electronically excited molecular nitrogen. The first part of this paper concerns the measurement of species generation efficiencies. Measurements of the concentration profiles of N,(B), N,(A), NF(a), and NF(b) produced in a H or D NF2 system are described. The H atoms were generated either in a microwave discharge of H2 in H e or from the F + H2 reaction. The D atoms were generated under D,-free conditions by the reaction F + Dz DF + D under the appropriate stoichiometric conditions. From the Nz(B) profiles it was possible to extract the efficiency of Nz(B) production in the H N F 2 system. The efficiency of N,(B) production also represents the minimum efficiency of N,(A) production. This statement is valid since the dominant loss channel for Nz(B) is radiative decay to the N,(A) state (radiative lifetimes, 7,,d, for the v levels of interest lie in the range 4-8 ps). Electronic quenching of the B state can be neglected under the typical reaction conditions employed in these studies since the total concentration of potentially efficient quenchers never exceeds lOI4
+
-
+
-
0 1988 American Chemical Society
5134
The Journal of Physical Chemistry, Vol. 92, No. 18, 1988
~ m - The ~ , N2(A) profiles indicate the existence of a significant quenching channel for N,(A). The NF(a) and NF(b) profiles provide additional data for testing and extending the currently proposed kinetic model of the H + NF2 system. Optical emission spectroscopy (OES) was used to measure the population profiles of the above species. By comparison of the emission intensity of the species of interest to that from the 0 + N O system, the absolute concentrations of each were determined.6-10 The emissions from N,(B), NF(a), and NF(b) all occur in regions of significant 0 + N O emission, allowing a direct comparison. However, the wavelengths of N2(A) emission are shorter than the onset of the 0 N O emission. An additional measurement of the relative sensitivity of the optical detection system provides the data required to determine N,(A) populations. At the end of this paper is a description of the estimation of the rate constant k, for the reaction of N(2D) atoms with NF(a) radicals, reaction 3. This reaction is responsible for the direct NF2 production of electronically excited nitrogen in the H system. The rate of production of N,(B,W) is
Hovis et al. I
"
-
-
( 6 ) Fontijn, A.; Meyer, C . B.; Schiff, H. I. J . Chem. Phys. 1964, 40, 64. (7) Vanpee, M.; Hill, K. D.; Kineyko, W. R. AIAA J . 1971, 9, 135. (8) Golde, M. F.; Roche, A. E.; Kaufman, F. J . Chem. Phys. 1973, 59, 3953. (9) Woolsey, G. A.; Lee, P. H.; Slater, W. D. J . Chem. Phys. 1977, 67, 1220. (10) Sutoh, M.; Morioka, Y.; Nakamura, M. J . Chem. Phys. 1980,72,20. (1 1) Lofthus, A.; Krupenie, P. H . J . Phys. Chem. ReJ Datu 1977, 6, 113.
0 IC
Thermo couple
NO2
N2F,/He injector opposite observation port
Figure 1. Flow tube used for optical emission studies.
d[N2(B9W)1/dt = kp[N(2D)1[NF(a)l - [N2(B)l/7rB (5)
Experimental Section Three different flow systems were used to obtain the results reported here. The optical emission measurements of the electronically excited molecular-state population profiles were done in a 7.6-cm diameter, high-velocity flow system. The measurement of k , which required the use of atomic resonance for determining [N(sD)], was done in a 2.5-cm diameter, high-velocity flow system that incorporated a vacuum monochromator detection system. The electron paramagnetic resonance (EPR) measurements of NF(a) populations were done in a 2.5-cm diameter flow tube that passed through the magnet of the EPR spectrometer. These systems will be described in more detail in subsequent sections. The gases used and their purities were as follows: He, Union Carbide, 99.995%; Ar, Union Carbide, 99.997%; N2, Air Products UHP, 99.99%; H2, Matheson UHP, 99.999%; D2, Air Products research grade, 99.99%; 02,Matheson UHP, 99.99%; NO, Matheson CP, 99.0%; NO2, Matheson CP, 99.5%; NZF4,95% N2F4 and 5% N,. ( a ) System for Population Profile Measurements. The flow system used to measure the population profile is illustrated in Figure 1. The flow tube is made of 7.6-cm diameter Pyrex tubing coated with halocarbon wax and is pumped through 20-cm diameter PVC pipe by a 940 L/s Stokes Model 61 5 blower backed by a 140 L/s Stokes Model 41 2 Microvac pump. The chemical trap consists of an 1-cm layer of soda lime on top of an 1-cm layer of NaCI.
'Observation port
Pressure measurement ports
+
+
To
I
I/I
+
where [N(2D)], [NF(a)], [N,(B)], and [N,(B,W)] are the absolute concentrations of N(2D) atoms, NF(alA) radicals, N2(B) molecules, and the N2(B) plus N2(W) manifold, respectively, and 7,B is the effective radiative lifetime of the total N2(B) concentration. If [N2(B)] is plotted as a function of time, a maximum occurs when d[N2(B,W)]/dt = 0. By making measurements of [N2(B)], [NF(a)], and [N2(B)] at the maximum, one can determine the rate constant, k,, by use of eq 5. This equation is derived from the assumptions that reaction 3 is the only process generating N2(B,W) in the H N F 2 flame, that radiative losses are the predominant loss channel for N,(B), and that N2(B) and N2(W) are collisionally coupled at a rate that is rapid compared with their production rate. It is important to appreciate that the loss term [N2(B)]/7,* uses only the N,(B) concentration since radiative losses of N2(W) are negligible.]'
.
N2F,/He
I
CllalS
/
/
,Thermocouole
connector
Heater
attached to outer ' surface of heater ~
threaded
Quartz heater
element?
n
I
5 5 cnl
~
element
4
symmetric holes ( - O S m m diaml
SUPPI)
Figure 2. Typical heated injector design.
The N2F4/Hemixtures used to generate the NF2 were injected through a moveable heated injector made of either Pyrex (operating at -230 "C) or stainless steel (operating at -200 "C). A detailed design of a typical injector is given in Figure 2. The power was supplied to the injector heating element for at least 2 h before any experiment to ensure thermal equilibrium throughout the heated section of the injector. Mass spectra of the effluent of a representative injector were taken with an on-line modulated-beam mass spectrometer to check that N2F4 dissociation was complete. Disappearance of the N2F4 parent peak upon heating indicated >95% dissociation. The N2F4/Hemixtures were prepared in an N2F4remote-handling facility. A thermocouple located 5 cm downstream of the observation ports was used to monitor gas temperatures. A number of ports were available for measuring pressures or injecting reagent gases such as NO or NO2. Pressures were measured with a Baratron capacitance manometer. Tylan Model FC260 flow controllers were used for measuring gas flow rates. H, 0, or F atoms were generated in a microwave discharge through mixtures of He and H2, 02,or F2 at a fixed inlet port located 53 cm upstream of the observation ports. N2(B), NF(a), and NF(b) profiles in the H NF, flame were determined under the following range of initial conditions: H 2 (( 1.3-5.6) X 10l8 molecules s-'); He carrier through the microwave discharge ((5.8-6.7) X 1020molecules s-l); -1% N2F4 in He ((2.1-6.1) X 1019 molecules s-l) through the heated injector. These flows resulted in a total pressure in the range of 80-90 Pa and flow velocities in the range of 6.3-7.4 m s-l. Additional measurements of peak [NF(a)] were performed under similar flow conditions with N2F4 mixtures in He that were 10% and 50% N2F4. The measurements with F + D2 as the D-atom source were performed under similar conditions except that the H2 flowing through the discharge was replaced by 5.7 X l O I 9 molecules s-' of a 10% F, in He mixture, and 6.2 X 1OIs molecules s-l of D2 was added -5 cm downstream of the discharge. When 0 atoms are needed for the 0 + N O standard emission, the H2 flowing through the discharge was replaced by a flow of 6.6 X I O l 9 molecules s-I of 02.The 0 concentration was determined by titration with NO,.',
+
(12) Clyne, M. A. A.; Thrush, B. A. Proc. R . SOC.London, A 1962,269, 404.
Excited N 2 in the H(D)
+ NF2 Flame
For the N2(A) profiles, an additional dc discharge source for N2(A) could be added 38 cm upstream of the observation ports. This source utilized the well-established technique of generating Ar metastables in a hollow-electrode dc discharge with subsequent transfer to N 2 to generate N2(A).13 In all of the population profiles determined by OES, a PAR Model 1420 intensified diode array, cooled to -25 "C, was employed as the detector. Acquisition and analysis of the data were performed with a PAR Model 1640 OMA I11 console. The spectrometer used to disperse the emission for these studies was a Jarrel Ash Model 82-499 with a 1200-line/mm grating. The observation ports were Suprasil 11. For the NF(a) measurements, a Schott RG780 long-pass glass filter prevented interference from short wavelengths transmitted in second order. For determination of N2(B) population profiles, it was necessary to first determine the N2(B)vibrational distribution under the conditions of our measurements. All vibrational distribution measurements were performed on the H + N F 2 flame with a microwave discharge in H2/He as the H-atom source and with flow rates and total pressures in the ranges given above. The resolution required precluded the use of the O M A for these measurements. Therefore, scans of the N2(B-A), Au = 4, 3, and 2, visible emission bands were made with two different detection systems. The first system, providing higher resolution, incorporated a 0.5-m Jarrell Ash Model 82-020 monochromator to disperse the emission and a Hamamatsu R758 photomultiplier for detection. However, for most of the scans, a lower resolution, 0.25-m Jarrell Ash Model 82-410 monochromator and an RCA C31034 photomultiplier were used to record the spectra. Spectra of the N2(B-A), Av = 1, 0, and -1, bands in the 850-1250-nm region were obtained with the same 0.25-m monochromator and a Northcoast Model EO-8 17 liquid-nitrogen-cooled intrinsic Ge detector. An RG780 Schott-glass filter eliminated any shortwavelength emission that could be transmitted through the monochromator in second order. The relative spectral sensitivities needed to convert emission intensities into relative populations for the detection systems operating in the visible region were determined from spectra of the 0 N O reference emission. The measurements of the relative spectral sensitivity of the intensified diode-array-based detection system in the UV, needed for determining the N2(A) concentrations, were made with an Eppley Model EP standard lamp and a Model 6162B-SR regulated power supply. In the UV measurements, the concomitant intense visible emission from the lamp, scattered inside the spectrometer, interfered with the UV signal. This interference was corrected by running background spectra with long-pass glass filters with cutoff wavelengths longer than the wavelengths of interest. To put upper limits on the N2(A) populations generated in the H + NF2 flame, we recorded four background-corrected spectra with the extra N2(A) source added to an H + NF2 flame. For all four spectra the He flow through the microwave discharge was 1.5 X 1021atoms SKI,the Ar flow was 1.7 X 1021atoms s-I, and the N2 flow was 8.5 X lozomolecules s-l, yielding a total pressure of 364 Pa and a flow velocity of 9.7 m s-l. When H atoms were needed, an H 2 flow of 9.0 X 10'' molecules s-l was added to the He before the microwave discharge. A flow of 4.1 X lOI9 molecules s-l of 0.85% N2F4 in H e could be added, either cold (no NF2) or heated (to yield NF,), 14 cm upstream of the observation ports. ( b ) Atomic Absorption Flow Tube System. The flow tube apparatus used for the measurement of k,, with a resonance lamp, absorption cell, and monochromator system at the fixed observation point for monitoring N(2D) atoms, is schematically represented in Figure 3. Hydrogen atoms were generated upstream in the flow tube (in the side-arm microwave discharge) by a 2.45-GHz microwave discharge in either pure H2 or a dilute mixture of H2 in He (typically 1% H2 in He). The NFz radicals were generated by thermal decomposition of N2F4diluted in He. Mixture ratios ranged from 1:l to 1:200 N2F4:He. Thermal
+
(13) Clark, W. G.; Setser, D. W. J . Phys. Chem. 1980, 84, 2225.
The Journal of Physical Chemistry, Vol. 92, No. 18, 1988 5135 N2F4/He, Ar
side-arm Baratron Dressure transducer
1 IIII I
I Ill1 wfl
Thermocouple probe 7
- Resonance lamp axis
Figure 3. Flow tube for the N(*D)
+ NF(a) studies.
decomposition of the N2F4 was achieved in the flow tube by flowing the N2F,/He mixture through the heated injector. The power was supplied to the injector heating element for at least 2 h before any experiment to ensure thermal equilibrium throughout the heated section of the injector. The internal surface of the flow tube and the external surface of the heated injector were coated with halocarbon wax to minimize losses of H atoms on the walls. The details of this system have been presented previously. l 4 The optical emission measurements necessary for the estimate of k, were performed as in the population profile measurements. Optical emission from the flow tube at the fixed observation point was collected through a horizontally mounted, adjustable window port aligned perpendicular to both the flow tube longitudinal axis and the resonance absorption axis and typically positioned adjacent to the reactive stream at the fixed observation point (see Figure 3). In this position the window provided optimum light collection from the flow tube. The emission was transmitted to the entrance slits of the OMA monochromator through a 0.64-m fiber optic cable. All the optical emissions to be monitored in this experiment were in the region of 870 20 nm. A long-pass filter (Schott-glass filter RG780) was placed between the fiber optic cable and the entrance slits to minimize the problem of short-wavelength emissions in second order and background-scattered light. The absolute concentrations of NF(a) and N2(B) were determined by using the OMA system as described above. Oxygen atoms were generated by a microwave discharge through a mixture of O2in carrier gas, in the discharge side arm. NOz was used without further purification to titrate the 0 atoms generated in the discharge and to generate the reference 0 + N O emission. (c) EPR Spectral Measurements. To double check the measurement of NF(a) populations by OES in the 7.6-cm diameter flow system, we made a separate set of measurements in which electron paramagnetic resonance was used to calibrate an optical detection system. The flow tube for these measurements was a halocarbon-wax-coated, 2.5-cm diameter Suprasil quartz tube. The flow conditions and densities were similar to those used for the previously described population profiles. The EPR system has been described previously.l5.l6 The optical calibration was performed in the same manner as described for O,(alA) except
*
(14) Whitefield, P. D.; Hovis, F. E. Chem. Phys. Lett. 1987, 135, 454. (15) Lilenfeld, H. V.; Richardson, R. J.; Hovis, F. E. J . Chem. Phys. 1981, 74, 2129. (16) Lilenfeld, H. V.; Carr, P. A. G.; Hovis, F. E. J . Chem. Phys. 1984, 81, 5730.
5136
Hovis et al.
The Journal of Physical Chemistry, Vol. 92, No. 18, 1988
that a Hamamatsu R636 photomultiplier tube was used instead of an intrinsic Ge detector.16 Two different H sources were used for these measurements. One source was the standard microwave discharge in H2/He mixtures, used for most of the NF(a) profiles described above. For the second source, F atoms were generated in a microwave of 10% F2 in He. H 2 was subsequently added to yield H atoms by the fast F + H2 reaction. This source had the advantage of producing H in the absence of H2.
Data Analysis (a) Optical Emission Calibration. The emission from 0 + N O recombination provides a reasonably well-established reference for determining the concentration of radiating species.6-10 For pressures above -40 Pa, the intensity of the 0 N O emission is given by
+
Ir(X)
= kr(h) [Ol [NO1
(6)
where Ir(h) is the absolute intensity at the wavelength X and kr(X) is the rate constant relating the intensity to the 0 and N O concentrations. From eq 6 it can be shown that the concentration of an emitting species M can be calculated from Sm,the measured intensity of M emission, and S,, the measured 0 N O reference emission intensity by the expression
+
TABLE I: N,(B-A) Transitions Used To Determine Relative Vibrational Populations u‘ u“ h . n m 104A,,c,8r,s-’ u‘ u” h , n m ~O-“A,+,,,~,S-’ 12 I1 10 9
8 7 6 11 10 9 8
7 6 5
8 7 6 5 4 3 2 8 7 6 5 4 3 2
574 579 584 589 595 600 606 624 631 638 645 653 661 669
6.49 6.11 5.38 4.38 3.25 2.14 1.19 4.14 5.86 7.31 8.15 8.10 7.09 5.26
7 6 5 4 3 2 1 5 4 3 2 0 1 0
5 4 3 2 1 0 0 5 4 3 2 0 2 1
715 726 737 748 761 773 888 917 940 965 991 1047 1189 1232
N2(B,v’=i-A,u’’) and N2(B,u’=j-+A,u’~bands, Ai and Aj are the Einstein A coefficients for those transitions (see Table I), and Ni and Nj are the populations in Nz(B,u=i and j). Once the vibrational distribution of N2(B) has been established, T , is calculated from 12
1/7e
=
CCfu/~v)
([oI[NoI~~/A~)S(S~(X)/S~(X)) dh (7) wheref, is the fractional population in level v and a=O
[MI =
where A, is the effective radiative rate constant for the M transition observed, k, is taken to be constant over the range of integration, and the integral extends over the emission band to which A , applies. The most useful N2(A-X) emission lines for determining Nz(A) concentrations lie in the 250-350-nm region.l3 Since the 0 N O reference emission used in the determination of Nz(B) concentrations is not a useful standard at C400 nm, another procedure must be used to measure N,(A) concentration. In this case the relative sensitivity of the optical detection system from 280 to 550 nm was determined by use of a standard lamp. A calibration factor relating absolute emission intensity to measured signal was determined at 546 nm by use of the 0 + N O emission with known [ O ]and [NO]. The relative sensitivities from the standard lamp measurements and the calibration factor at 546 nm were used to calculate the calibration factor relating 276-nm N2(A,u=O) Nz(X,u=6) emission intensity to N2(A,u=O) population. (b) N2(B) Vibrational Distribution and Production Efficiencies. The efficiency of N2(B) production EfB is defined as
+
-
EfB = total [N,(B)] produced/total [N2F4]consumed
(8)
Since radiative losses are the dominant loss channel for N,(B) in these studies of the H + NF2 system total [N2(B)] produced = total [Nz(B)]lost = Jff([Nz(B)tI/T~) dt (9) where tf is the time for the reactions to go to completion and T , is the effective radiative lifetime for the N2(B) vibrational distribution produced. To use eq 9, one must measure the Nz(B) vibrational distribution so that T , can be determined. The N2(B,u”+A,u’’) bands with Av = u’- u” = 4, 3, and 2 occur in the 560-800-nm region and are useful for determining the relative vibrational distribution for u = 2 to u = 12.” For determination of the u = 0 and v = 1 populations, it is necessary to observe the N2(B-+A), Au = 1, 0, and -1, emission bands in the 850-1250-nm region. For both of the spectral regions, the 0 + NO emission is used to correct for differences in the relative sensitivity of the detection system. The relative populations of the N,(B) vibrational levels can be calculated from the emission spectra by use of the equation Ii/Ij
= AiNi/AjNj
(10)
where Ii and Ij are the integrated intensities of the
2.03 4.38 6.86 8.36 7.73 4.44 8.72 1.91 2.94 2.85 1.25 6.25 1.85 3.56
(1 1)
T“ is the radiative lifetime of level u. Using T,, the Nz(B) population profiles, and eq 8 and 9, one can calculate the N2(B) production efficiencies. (c) N 2 ( B ) Populations. The concentration of N2(B) at a particular time t is needed to calculate the total N,(B) produced. This was done by comparing the N2(B) emission intensity for a particular band or group of bands to that from the reference 0 NO emission, as described above. In our measurements, three different portions of the N2(B-A) emission were used to determine [N2(B)]values: (1) the Au = 4 band from u’ = 12 6, (2) the Au = 3 band from u ‘ = 11 5, and (3) the u ’ = 2 v’’ = 1 line. For each of the bands the effective radiative rate constant needed for eq 7 is given by
+
--
-
A, = Efv,Aur
(12)
where Ad is the radiative rate constant for the v’- u”component of the band and the sum extends over all d i n the band. The value for f,,is calculated from ftiJ
= Not/ENvI
(13)
with the sum extending over all u’contributing to the observed band. For the Au = 3 and 4 bands, k,(X) is taken to be 1.4 X cm3 SKI nm-’, the average of the values reported in ref 4, 5,7, and 8. For the u’ = 2 v’’ = 1 band at 874 nm, the average cm3 s-l nm-I. The values for the A”, value for k,(X) is 9 X were taken from ref 11. Note that the u ’ = 2 u ” = 1 band is overlapped by the NF(a-X) emission. The deconvolution of these bands will be described in detail in part d. (d)NF(a) Populations. For accurate determination of NF(a) concentrations from a X emission at 874 nm, it was necessary to correct only for the contribution from N,(B-A), u’ = 2 c” = 1, emission since no H F overtone emission was observed under typical reaction conditions. An empirical technique was employed. A pure NF(a-X) spectrum was recorded from the H + NF2 flame with [NF,], >> [HI, and the [NF(a)] was determined as .described above. The value of the 0 NO radiative rate constant at 860-880 nm was taken to be 9 X cm3 s-l nm-1:,7*9,10and X transition was taken the radiative rate constant for the a to be 0.18 S-I.~ A pure N,(B-A), v’ = 2 u” = 1, reference spectrum was recorded from a microwave discharge in N2/Ar, and the N2(B,v=2) concentration was determined by using a radiative rate constant for the transition of 6.17 X lo4 s-lS1l Then, a least-squares analysis was used to find the linear combination of the NF(a) and N2(B) reference spectra that provided a best fit to the unknown mixed spectrum. This method has the advantage that both [NF(a)] and [N2(B,u=2)] are determined. Since the N,(B) vibrational distribution was previously determined,
-
-
-
-
+
--
Excited N 2 in the H(D)
+ NF2 Flame
The Journal of Physical Chemistry, Vol. 92, No. 18, 1988 5137
TABLE 11: Peak NF(a) Concentrations opt calibr H-atom 10-13[NF2]o, [NF(a)lPk/ techn source cm-’ [NFZIO a I .3 0.16 O+NO a 3.7 0.14 O+NO a 3.7 0.19 O+NO
