T. Amano, P.F. Bernath and A.R.W. McKellar- Direct Observation of the nu-1 and nu-3 Fundamental...

8/3/2019 T. Amano, P.F. Bernath and A.R.W. McKellar- Direct Observation of the nu-1 and nu-3 Fundamental Bands of NH2 b

1/14

JOURNAL OF MOLECULAR SPECTROSCOPY 94, loo- 113 ( 1982)

Direct Observation of the vl and v3 Fundamental Bands of NH2 byDifference Frequency Laser Spectroscopy

T. AMANO, . F. BERNATH, AND A. R. W. MCKELLARHerzberg Institute of Astrophysics, National Research Council of Canada, Ottawa,

Ontario KIA OR6 CanadaThe Y, and vj bands of the NH2 radical were detected in absorption in the 2.9- to 3.2~pm

region using a tunable difference frequency laser and a long-path Zeeman-modulated dischargecell. About 100 rotation-vibration transitions were measured and a simultaneous analysis ofthe Coriolis-coupled Y, and Y, states was made. It was found that the Y, band is considerablystronger than Y,, in contrast to the similar molecule H20. These results may prove useful ina search for interstellar NH2 by means of its rotation-vibration spectrum.

I. INTRODUCTIONThe NH* radical has been the subject of extensive high-resolution spectroscopic

analysis, particularly by means of its k2A1-z2BI electronic transition in the visibleregion (Z-6), but studies of its infrared spectrum have been relatively limited. In1965 two features were observed by Milligan and Jacox (7) in the low-temperaturematrix isolation spectrum of photolyzed NH3 at 1499 and 3220 cm-, which wereassigned by them as the u2 (bend) and v3 (asymmetric stretch) vibrations, respec-tively, of NH*. Recently, the v2 band in the gas phase has been observed andanalyzed using laser magnetic resonance spectroscopy (8). In addition to directinfrared observations, information on the excited vibrational states of NH2 also hasbeen obtained from the electronic spectrum. Kroll (4) obtained a vibrational fre-quency of 3221 cm- from an analysis of laser-induced fluorescence data and con-cluded that it was the vI (symmetric stretch) vibration rather than u3, because ofthe observed selection rules. By detailed analyses of the visible emission spectrum,Birss et al. (5) have studied the (u~LJ~u~) = (020) excited bending state and veryrecently Vervloet and Merienne-Lafore (6) have obtained an improved value ofv, = 3219.36 cm- and a tentative value of v3 = 3280 cm-. Their detection of levelsin the u3 = 1 state was made possible by Coriolis perturbations between the (100)and (001) states.In the present paper, we report the direct detection of the vI and v3 fundamentalbands of NH2 in the 2.9- and 3.2-pm region of the infrared. About 100 rotation-vibration transitions have been measured and a detailed analysis of the Coriolis-coupled ( 100) and (001) states has been made. Our results confirm, and consid-erably extend, those of Vervloet and Merienne-Lafore (6). The observations weremade using a tunable difference frequency laser system and a long pathlengthZeeman-modulated discharge absorption cell. The difference frequency laser wasdeveloped as a high-resolution spectroscopic tool by Pine (9). Our laser system was0022-2852/82/070100-14$02.00/OCopyright 0 1982 by Academic Press. Inc.All rights of reproduction n any form rescrvcd.

100


2/14

Y, AND yj BANDS OF NH2 101constructed by Oka (IO) for use in studying unstable molecules and the detectionof the spectrum of Hz by Oka (I 1) confirmed the value of the technique for suchtransient species. Vibrational spectra of two further ions, HeH+ (Z2) and NeH+(13), have since been studied using Okas original apparatus. By adding Zeemanmodulation to an absorption cell similar to that used for these ions, a number ofparamagnetic free radicals have been detected in the 3-pm region, of which NH2was the first. Similar studies of free radicals have been made in other laboratoriesusing tunable infrared diode lasers (14) and color center lasers (15).

II. EXPERIMENTAL DETAILSThe apparatus used in this investigation is shown schematically in Fig. 1. The

difference frequency laser system was almost identical to that of Oka (10, II).Visible radiation from a cw single-mode tunable dye laser is mixed with that froma single-mode Ar+ laser in a temperature-controlled lithium niobate crystal. Theresulting difference frequency lies in the infrared and may be tuned over a rangeof about 2.2 to 4.4 pm with a power of a few microwatts. The infrared linewidthis governed by those of the two visible lasers and was of the order of 10 MHz.

The discharge absorption cell used here was similar to that used by Oka forH: (I I) with the addition of provision for Zeeman modulation, It was made froma 2.4-cm-diam. Pyrex tube fitted with cylindrical electrodes and Brewster angle endwindows. Multiple-reflection mirrors were located outside the cell adjacent to thecell windows and the cell was wrapped with a two layer solenoidal modulation coilwith about 8.5 turns/cm. The modulated field length was 1 m, the discharge length

Ar + Pump Laser

~~~,FIG. 1. Schematic diagram of the apparatus used in this investigation.


3/14

102 AMANO, BERN ATH, AND MCKEL LAR1.5 m, and the mirror separation 2 m. A sinusoidal modulation current at 3.5 kHzwas applied to the coil along with a variable dc bias to achieve zero-based fieldmodulation. The maximum current of about 20 A gave a modulation amplitudeof about 200 G (peak to peak). The modulation current, from a high-power audioamplifier, and the bias current were matched to the Zeeman coil with a transformerand tuned LC series circuit as shown in Fig. 1. The cell and coil were cooled byimmersion in either ice or dry ice baths.The infrared laser beam was passed through the cell 16 times for a total effectivemodulated path of about 16 m. Part of the infrared beam was directed througha reference gas cell for wavelength calibration purposes (see Fig. 1) and eachinfrared beam was monitored by a liquid nitrogen cooled InSb detector. In additionto the Zeeman modulation, we also employed simultaneous frequency modulationof the infrared radiation at 2.5 kHz in order to record the reference gas spectrum.The spectra of CzH4 (16) and NzO (17) were used for reference and the accuracyof the NH2 measurements is estimated to be about 0.003 cm-.

NH2 was produced by a dc discharge in flowing NH3 at a pressure of about 0.8Torr and a discharge current of about 30 mA. Before searching for the infraredspectrum, we monitored the visible laser-induced fluorescence from NH2; this ob-servation, and the infrared study, showed that the NH2 production was not criticallydependent on the pressure, current or pumping speed.

III. RESULTS AND ANALYSISThe term values obtained by Vervloet and Merienne-Lafore (6) for the u, = 1state were used to predict transition frequencies for our initial searches. After a

number of vI band lines were found, some strongly perturbed v3 band lines werealso detected on the basis of Ref. (6). Analysis of these preliminary infrared mea-surements resulted in more accurate and extensive predictions for both bands andultimately 59 rotation-vibration transitions were identified in vI and 37 in v3; themeasurements are listed, together with their assignments, in Tables I and II. Foreach such transition we generally measured two or more fine-structure components.Examples of v, and v3 band transitions detected using the Zeeman-modulationtechnique are shown in Figs. 2 and 3, respectively. The vI band transitions weregenerally much stronger than those of v3 as is discussed in more detail below. Amore detailed look at one part of the absorption spectrum of discharged ammoniais shown in Fig. 4. The upper trace in Fig. 4 is a conventional transmission spectrumrecorded by chopping the infrared beam followed by phase-sensitive detection. Twoprominent NH3 lines are evident in this trace. Much greater sensitivity is gainedby using frequency modulation as shown in the two middle traces of Fig. 4. Withthe discharge on, new lines appear in the spectrum, some of which are probablydue to hot NH3. However, three of these new features are due to NH*, as shownby the lowest trace, which was taken using Zeeman modulation. The region of Fig.4 is relatively free of NH, absorption compared to much of the spectrum that wesurveyed and Zeeman modulation was useful for identification but not essential fordetection. However, in many other regions the Zeeman modulation was essentialfor the detection of the NH2 transitions.


4/14

Y, AND Y, BANDS OF NH2TABLE I

Observed Transitions in the Y, Fundamental Band of NH2Assignment Observed o-c

103

N W, J (CIn-- -1(cm )6 0 6 + 6 1 56 0 6 ~ 6 1 54 3 1 + 4 4 04 3 1 - 4 4 01 1 1 + 2 2 01 1 1 - 2 2 01 1 1 * 2 2 01 1 0 + 2 2 11 1 0 + 2 2 14 2 3 * 4 3 24 2 3 , 4 3 23 1 3 * 4 0 43 1 3 * 4 0 43 2 2 - 3 3 13 2 2 ' 3 3 12 0 2 - 3 1 32 0 2 - 3 1 32 0 2 , 3 1 34 2 2 + 4 3 14 2 2 ~ 4 3 13 1 3 - 3 2 23 1 3 * 3 2 21 0 1 - 2 1 21 0 1 - 2 1 22 1 2 - 2 2 12 1 2 * 2 2 14 1 3 * 4 2 24 1 3 * 4 2 22 1 1 - 2 2 02 1 1 + 2 2 02 1 1 - 2 2 03 0 3 ~ 3 1 23 0 3 - 3 1 23 1 2 ~ 3 2 13 1 2 - 3 2 10 0 0 - 1 1 10 0 0 * 1 1 1I l l - 2 0 2I l l - 2 0 21 1 1 + 2 0 22 0 2 - 2 1 12 0 2 * 2 1 12 0 2 - 2 1 12 0 2 + 2 1 11 0 1 + 1 1 01 0 1 + 1 1 01 0 1 - 1 1 01 0 1 * 1 1 02 2 0 - 3 1 32 2 0 - 3 1 3l l O* l Ol1 1 0 - 1 0 1l l O* l Ol

5 . 5 + 5 . 56 . 5 * 6 . 53 . 5 + 3 . 54 . 5 * 4 . 5i : : * 1 . 5. 51 . 5 + 2 . 50 . 5 f 1 . 51 . 5 * 2 . 53 . 5 + 3 . 54 . 5 c 4 . 53 . 5 * 4 . 52 . 5 * 3 . 52 . 5 * 2 . 53 : : f 3 . 5. 51 . 5 * 2 . 52 . 5 + 3 . 53 . 5 + 3 . 54 . 5 + 4 . 52 . 5 + 2 . 53 - 5 f 3 . 5c . 5 * 1 . 51 . 5 + 2 . 51 . 5 + 1 . 52 . 5 * 2 . 53 . 5 + 3 . 54 . 5 + 4 . 52 . 5 * 1 . 51 . 5 + 1 . 52 . 5 * 2 . 55 : ; * 3 . 5. 52 . 5 * 2 . 5Z + 0 . 5. 50 . 5 * 1 . 51 . 5 * 1 . 51 . 5 + 2 . 50 . 5 * 1 . 52 . 5 + 1 . 5: : i : * 2 . 5. 51 . 5 * 2 . 51 . 5 * 0 . 50 . 5 * 0 . 5: : ' 5 + 1 . 5. 52 . 5 * 3 . 51 . 5 * 2 . 51 . 5 * c . 51 . 5 * 1 . 50 . 5 * 0 . 5

3 1 2 6 . 2 1 1 0.0003 1 2 6 . 3 7 3 - 0 . 0 0 23 1 2 6 . 1 4 7 - 0 . 0 0 6 a3 1 2 6 . 6 0 0 0 . 0 0 1 =3 1 3 3 . 2 3 3 0 . 0 0 23 1 3 3 . 4 5 3 - 0 . 0 0 03 1 3 3 . 7 5 7 - 0 . 0 0 13 1 3 9 . 4 9 5 0 . 0 0 23 1 3 9 . 7 6 7 0 . 0 0 13 1 4 5 . 8 0 3 0 . 0 0 73 1 4 6 . 1 2 2 0 . 0 0 73 1 4 6 . 0 3 5 0 . 0 0 23 1 4 6 . 0 5 7 - 0 . 0 0 13 1 5 0 . 2 9 3 - 0 . 0 0 23 1 5 0 . 6 9 6 - 0 . 3 0 13 1 5 5 . 5 1 9 - o . o o c3 1 5 5 . 5 7 7 0 . 0 0 13 1 5 5 . 6 4 7 0 . 0 0 53 1 5 8 . 8 9 4 0 . 0 0 53 1 5 9 . 2 4 3 0 . 0 0 93 1 6 2 . 6 3 7 o . o c 23 1 6 2 . 9 2 5 - 0 . 0 0 03 1 7 0 . 7 6 5 0 . 0 0 23 1 7 0 . 8 8 0 0 . 0 0 43 1 7 1 . 4 6 6 - 0 . 0 0 43 1 7 1 . 8 5 4 - 0 . 0 0 23 1 8 1 . 4 4 5 - 0 . 0 0 33 1 8 1 . 5 9 0 - 0 . 0 0 43 1 8 4 . 2 5 9 - 9 . 0 0 03 1 8 4 . 4 5 9 0 . 0 0 23 1 8 4 . 7 8 6 - 0 . 0 0 03 1 8 4 . 8 6 4 0 . 0 0 03 1 8 4 . 9 9 0 0 . 0 0 33 1 8 5 . 0 2 0 0 . 0 0 33 1 8 5 . 2 3 5 0 . 0 0 63 1 8 7 . 3 6 6 0 . 0 0 13 1 8 7 . 5 9 8 0 . 0 0 23 1 8 8 . 3 2 8 - 0 . 0 0 03 1 8 8 . 3 8 6 0 . 0 0 13 1 8 8 . 5 4 6 - 0 . 0 0 53 1 9 6 . 6 3 3 0 . 0 0 33 1 9 6 . 6 8 8 0 . 0 0 13 1 9 6 . 8 3 1 - 0 . 0 0 33 1 a 6 . 8 8 9 - 0 . 0 0 23 2 0 3 . 3 4 3 0 . 0 0 43 2 0 3 . 3 7 6 0 . 0 0 23 2 0 3 . 6 0 5 0 . 0 0 03 2 0 3 . 6 3 8 - 0 . 0 0 13 2 0 8 . 7 6 0 - 0 . 0 0 03 2 0 9 . 1 4 6 0 . 0 0 03 2 3 4 . 0 1 8 - 0 . c o 53 2 3 4 . 0 5 2 - 0 . 0 0 43 2 3 4 . 2 7 6 - 0 . 0 0 2

'These less accurately measured transitions were givenreduced weight (0.1) in the least-squares fit.

b These uncertain measurements were omitted from theleast-squares fit.

The lineshapes exhibited by the Zeeman-modulated traces in Figs. 2-4 are func-tions of the modulation amplitude and the spin-splittings of the energy levels in-volved in each transition. Identification of the true line center was usually fairly


5/14

104 AMANO, BERN ATH, AND MCKEL LARTABLE I-Continued

Assignment Observed O-CN KaKc J (cm- (cm-1 1 0 + 1 0 1 0 . 5 * 7 . 5 3 2 3 4 . 3 0 9 - 0 . 0 0 32 1 1 - 2 0 22 1 1 - 2 0 22 1 1 - 2 0 22 1 1 - 2 0 23 2 1 - 3 1 23 2 1 - 3 1 24 2 2 * 4 1 34 2 2 + 4 133 1 2 - 3 0 3

3 1 2 * 3 0 33 1 2 + 3 0 33 1 2 - 3 0 32 2 0 - 2 1 12 2 0 - 2 1 12 2 0 - 2 1 12 2 0 * 2 1 11 1 1 - 0 0 01 1 1 - 0 0 05 2 3 + 5 1 45 2 3 - 5 1 42 2 1 - 2 1 22 2 1 + 2 1 24 1 3 . 4 0 44 1 3 * 4 0 42 1 2 - 1 0 12 1 2 - 1 0 12 1 2 * 1 0 ' 14 3 1 + 4 2 24 3 1 + 4 2 23 0 3 - 2 1 23 0 3 + 2 1 23 0 3 - 2 1 23 2 2 - 3 1 33 2 2 + 3 1 33 3 0 - 3 2 13 3 0 - 3 2 13 3 1 - 3 2 23 3 1 * 3 2 24 2 3 , 4 1 44 2 3 * 4 1 43 1 3 , 2 0 23 1 3 - 2 0 23 1 3 - 2 0 24 1 3 + 3 2 24 1 3 + 3 2 24 3 2 - 4 2 34 3 2 - 4 2 35 3 3 - 5 2 45 3 3 + 5 2 44 0 4 + 3 1 34 0 4 . 3 1 35 2 4 + 5 1 5

2 . 5 * 1 . 5: : : + 2 . 5. 51 . 5 * 2 . 53 . 5 + 3 . 52 . 5 * 2 . 54 . 5 + 4 . 53 . 5 + 3 . 53 . 5 + 2 . 5: : : + 3 . 5. 52 . 5 + 3 . 5f : : * 2 . 5. 51 . 5 + 1 . 51 . 5 + 2 . 5i : : + 0 . 5. 55 . 5 + 5 . 53 : : * 4 . 5. 51 . 5 + 1 . 54 . 5 + 4 . 53 . 5 + 3 . 52 . 5 * 1 . 51 . 5 + 0 . 51 . 5 * 1 . 5: : : + 4 . 5. 52 . 5 + 1 . 53 . 5 + 2 . 53 2 : : * 2 . 5. 52 . 5 * 2 . 53 . 5 * 3 . 52 . 5 + 2 . 53 . 5 f 3 . 52 . 5 * 2 . 54 . 5 + 4 . 5Z : : . + 3 . 5. 52 . 5 + 1 . 53 2 : : * 2 . 5. 54 . 5 + 3 . 54 . 5 + 4 . 52 : : + 3 . 5. 53 u : : + 4 . 5. 54 . 5 + 3 . 55 . 5 + 5 . 5

3239.358 0 . 0 0 13 2 3 9 . 4 1 5 0 . 0 0 13 2 3 9 . 5 5 3 - 0 . 0 0 23 2 3 9 . 6 1 1 - O. OGl3 2 4 7 . 4 0 0 - 0 . 0 0 13 2 4 7 . 5 9 2 - 0 . 0 0 13 2 4 7 . 6 5 8 0 . 0 0 83 2 4 7 . 7 6 4 0 . 0 0 73 2 4 9 . 0 1 6 - 0 . 0 0 43 2 4 9 . 0 9 5 - 0 . 0 0 53 2 4 9 . 2 1 7 - 0 . 0 0 13 2 9 9 . 2 9 6 - 0 . 0 0 23 2 4 9 . 7 4 6 - 0 . 0 0 23 2 4 9 . 9 5 0 - 0 . 0 0 23 2 5 0 . 2 5 7 0 . 0 0 13 2 5 0 . 4 6 2 0 . 0 0 23 2 5 0 . 4 2 9 o . o o c3 2 5 0 . 6 5 6 0 . 0 0 53 2 5 4 . 0 3 1 - 0 . 0 0 43 2 5 U. 1 3 0 - 0 . 0 0 13 2 6 2 . 9 3 9 - 0 . 9 0 13 2 6 3 . 3 0 1 - 0 . 0 0 13 2 6 3 . 1 4 1 0 . 0 0 03 2 6 3 . 2 6 3 - 0 . 0 0 03 2 6 6 . 1 U8 0 . 0 0 23 2 6 6 . 2 5 4 - 0 . 0 0 13 2 6 6 . 2 9 0 0 . 0 0 13 2 6 6 . 7 0 7 - 0 . 0 0 03 2 6 7 . 0 1 4 - 0 . 0 0 23 2 6 8 . 6 5 5 - 0 . 0 0 23 2 6 8 . 7 2 7 0 . 0 0 23 2 6 8 . 8 0 5 0 . 0 0 13 2 6 9 . 8 1 6 - 0 . 0 0 53 2 7 0 . P 9 6 0 . 0 0 13 2 7 3 . 2 5 4 0 . 0 0 53 2 7 3 . 6 2 0 0 . 0 0 13 2 7 8 . 9 2 2 0 . 0 0 23 2 7 9 . 2 9 5 0 . 0 0 13 2 7 8 . 9 6 3 o . o c 33 2 7 9 . 2 0 3 0 . 0 0 9 a3 2 7 9 . 6 7 4 - 0 . 0 0 33 2 7 9 . 7 4 0 0 . 0 0 03 2 7 9 . 7 9 0 - o . o o 7 b3 2 7 9 . 8 3 5 - 0 . w43 2 8 0 . 0 2 9 - 0 . 0 0 43 2 8 0 . 6 9 5 - 0 . 0 0 13 2 8 0 . 9 8 2 - 0 . 0 0 33 2 8 4 . 4 4 5 - 0 . 0 0 23 2 8 U. 6 8 2 - 0 . 0 0 33 2 8 6 . 7 3 6 - 0 . 0 0 53 2 8 6 . 7 6 7 - @. 0 0 23 2 9 0 . 0 5 4 0 . 0 5 5 b

clear but uncertainties in doing this sometimes contributed errors of the samemagnitude (-0.003 cm-) as our absolute measurement accuracy.The net electron spin, S = l/2, of NH2 in its B, ground electronic state resultsin a splitting of each rotational energy level, denoted by IVKaKonto two components.The resulting angular momentum is J = N + S, and in NH2 the level with J= N - l/2 (F2) is located above the level with J = N + l/2 (F,). Our analysiswas carried out using a doublet asymmetric rotor computer program originallydeveloped to analyze laser magnetic resonance spectra of Coriolis-coupled bands


6/14

Y, AND v3 BANDS OF NH2

TABLE I-ContinuedAssignment Ob s e r v e d o - c

N' L KcJ

k m- ' ) ( c m- ' )I 7 ( 1 * 5 1 5 O. ' l 6 3

105

4 ; 4 * 3 0 3SUl-5325 4 1 * 5 3 22 2 1 - 1 1 02 2 1 - 1 1 02 2 1 - 1 1 06 1 5 - 6 0 66 1 5 - 6 0 64 4 0 * 4 3 14 4 0 - 4 3 14 4 1 - 4 3 24 4 1 * 4 3 25 4 2 + 5 3 35 4 2 - 5 3 32 2 0 + 1 1 12 2 0 - 1 1 12 2 0 - 1 1 15 0 5 + 4 1 45 0 5 - 4 1 45 1 5 + 4 0 45 1 5 + 4 0 43 3 1 - 2 2 03 3 1 + 2 2 03 3 0 + 2 2 13 3 0 * 2 2 14 3 2 - 3 2 14 3 2 + 3 2 14 3 1 - 3 2 24 3 1 ~ 3 2 25 3 3 - 4 2 25 3 3 * 4 2 2

3 . 5 + 2 . 55 . 5 + 5 . 54 . 5 + 4 . 52 . 5 * 1 . 51 . 5 + 0 . 51 . 5 * 1 . 56 . 5 * 6 . 55 . 5 + 5 . 5i : : . * 4 . 5. 54 . 5 + 4 . 5; : i : * 5 . 5. 54 . 5 + 4 . 52 . 5 + 1 . 51 . 5 + 0 . 51 . 5 + 1 . 54 . 5 f 3 . 55 . 5 + 4 . 55 . 5 t 4 . 5: : : + 3 . 5. 52 . 5 * 1 . 53 . 5 * 2 . 52 . 5 * 1 . 54 . 5 * 3 . 53 . 5 * 2 . 54 . 5 + 3 . 53 . 5 + 2 . 55 . 5 * 4 . 54 . 5 * 3 . 5

3292.469 - 0 . 0 0 13 2 9 2 . 5 0 4 0 . 9 0 13 2 9 3 . 5 6 7 0.0013 2 9 3 . 8 2 5 0 . 0 0 63 2 9 5 . 6 6 5 - 0 . 0 0 33 2 9 5 . 9 1 2 - 0 . 9 0 13 2 9 6 . 1 1 7 - 0 . 0 0 13 2 9 6 . 7 4 1 - 0 . 9 0 13 2 9 6 . 9 0 9 0 . 9 0 53 2 9 1 . 1 7 7 - O . OOg a3 2 9 7 . 5 1 9 - 0 . 0 0 23 2 9 8 . 8 8 6 - 0 . 0 0 53 2 9 9 . 1 9 6 - 0 . 0 0 03 3 0 0 . 5 5 5 - 0 . 0 0 13 3 0 0 . 6 9 8 0 . 0 0 13 3 0 1 . 7 1 2 - 0 . 0 0 23 3 0 1 . 9 9 1 0 . 0 0 03 3 0 2 . 2 2 2 0 . 0 0 03 3 0 2 . 9 5 u - 0 . 0 0 1 13 3 0 2 . 9 6 6 - 0 . 0 0 23 3 0 5 . 5 6 2 0 . 0 0 63 3 0 5 . 5 8 2 0 . 0 0 93 3 4 1 . 6 4 6 0 . 0 0 23 3 4 1 . 3 0 5 0 . 0 0 33 3 4 2 . 4 6 8 0 . 0 0 43 3 4 2 . 1 1 8 0 . 0 0 23 3 5 8 . 2 4 7 - 0 . 0 0 13 3 5 8 . 4 8 4 - 0 . 0 0 23 3 6 5 . 1 8 6 - 0 . 0 0 03 3 6 5 . 4 0 5 - 0 . 0 0 23 3 7 1 . 8 8 4 - 0 . 0 0 13 3 7 2 . 0 8 9 - 0 . 0 0 4

(18). The A-reduced asymmetric rotor Hamiltonian of Watson (19) was used to-gether with the spin-rotation Hamiltonian of Brown and Sears (20) (see, for ex-ample, Eqs. ( 1) and (2) of (21)). It was necessary to analyze the vl and v3 banddata simultaneously because of the c-type Coriolis interaction between the (100)and (001) states. This interaction gives rise to an additional term in the rotationalHamiltonian connecting the two states given by

K,, = i&N, + z(N,Nb + Ndv,). (1)The resulting matrix elements are given by(u, J, N, klH,,,lu, J, IV, k + 1) = [*(l/2)& + Z(k + l/2)]

x [N(N + 1) - k(k f l)], (2)where u = (100) and u = (001). The constant f is related to the usual Corioliscoupling constant !: by

Ef3 = fi3cN~I/~3Y2 + b3/~W219 (3)and 2 is a higher-order term which is essentially a K-dependent correction to 4.Our 2 is equivalent to hd3 of Flaud and Camy-Peyret (22) and toF of Tanaka andMorino (23). The resulting molecular Hamiltonian was set up in a Wang-type


7/14


8/14

Y, AND v1 BANDS OF NH2 107TABLE II-Continued

Assignment Observed o-cNWC

J(cm-') wn-3 0 3 * 2 0 2 2 . 5 + 1 , s 3 3 5 8 . 2 3 8 0 . 0 0 43 2 2 * 2 2 1 3 . 5 + 2 . 5 3 3 5 9 . 8 3 0 - 0 . 0 0 23 2 2 - 2 2 1 : : : + 1 . 5 3 3 5 9 . 6 8 9 - 0 . 0 0 13 2 1 - 2 2 0 + 2 . 5 3 3 6 4 . 7 6 3 0 . 0 0 1

3 2 1 * 2 2 0 3 2 : : + 1 . 5 3 3 6 4 . 6 1 9 0 . 0 0 13 1 2 + 2 1 1 * 2 . 5 3 3 6 8 . 5 2 8 - 0 . 0 0 8 a3 1 2 - 2 1 1 2 . 5 + 1 . 5 3 3 6 8 . 5 2 8 - 0 . 0 0 1 a4 1 4 - 3 1 34 l 4 - 3 3 3 i : i : + 3 . 5 3 3 7 1 . 3 3 3 - o . o 3 S '* 2 . 5 3 3 7 1 . 3 5 7 - 0 . 0 3 3 b4 3 2 * 3 3 1 3 . 5 - 2 . 5 3 3 7 6 . 2 3 9 0 . 0 0 6 au 3 2 + 3 3 t 4 . 5 * 3 . 5 3 3 7 6 . 3 6 7 O. Ol P4 3 1 * 3 3 0 3 . 5 + 2 . 5 3 3 7 7 . 5 0 5 o . o o z a4 3 1 * 3 3 0 4 . 5 + 3 . 5 3 3 7 7 . 6 1 9 0 . 0 0 7 a4 2 3 ' 3 2 2 3 u : 5 5 . 3 . 5 3 3 7 9 . 3 2 0 - 0 . 0 0 24 2 3 + 3 2 2 + 2 . 5 3 3 7 9 . 2 6 0 - O. @Ol4 2 2 - 3 2 1 4 . 5 * 3 . 5 3 3 8 9 . 1 1 5 - 0 . 0 0 7 =4 2 2 * 3 2 1 3 . 5 * 2 . 5 3 3 8 9 . 0 5 3 - o . o o s a5 3 2 + 4 3 1 5 . 5 * 4 . 5 3 3 9 8 . 5 6 0 0 . 0 0 25 3 2 ' 4 3 1 4 . 5 - 3 . 5 3 3 9 8 . 6 1 2 - 0 . 0 0 4aThese less accurately measured transitions ere givenfeduced weight (0.1) in the least-squares it.

bThese uncertain measutement~ were omitted from theleast-squares it.

estimated experimental uncertainty. The Coriolis parameters & and 2 were highlycorrelated in the fit (correlation coefficient = -0.999958). A fit of virtually thesame quality could be obtained by fixing & equal to zero, as indeed was done byFlaud and Gamy-Peyret (22) in their analysis of the analogous bands of H,O, but

3133.6cmI .7 .6F I G . 2 . The lower trace shows the I,, - 220 transition in the Y, band of NH2 recorded using theZ&man-modulation technique. The upper trace is a spectrum of C,H., for wavelength calibration re-corded simultaneously using frequency modulation.


9/14

108 AMANO, BERNATH, AND MCKELLAR

3322*2cm- *I *O 3321*9cm-FIG. 3. The lower trace shows the loI + Ooo ransition in the Y)band of NH2 recorded using Zeemanmodulation. The upper trace, due to N20, is for wavelength calibration.

we feel that in the present case it is more realistic to allow both parameters to vary.All other correlation coefficients in the present fit had absolute values less than0.95 and all but three were less than 0.90.The (100) state parameters agree fairly well with those of the less precise de-termination by Vervloet and Merienne-Lafore (6). Many of the differences thatdo exist between the two sets of parameters may be ascribed to the fact that weexplicitly included the Coriolis interaction with (001) and allowed more higher-order parameters to vary. Our value for the y1 band origin is just 0.01 cm- abovethat of (6). However, our value of Q, 3301.110 cm-, is considerably different fromtheir estimate of 3280 cm- even though their analysis (6) of perturbed u3 statelevels agrees completely with ours.

IV. DISCUSSION AND CONCLUSIONSExcept for its unpaired electron spin, the structure of NH2 is quite similar tothat of Hz0 and the known v1 and v3 spectra of Hz0 (22) provided a useful guidefor the present analysis. The changes in the rotational constants of the two molecules


10/14

V, AND Y, BANDS OF NH2 109Spectroscopy of an NH, Discharge

FrequencyModulation(Discharge off

I-3250~7 cm

I11-000

FZ-FI

I3250.5 cm-

ZeemanModulationVI NH,

.L-3250.3 cm

FIG. 4. A smail portion of the absorption spectrum of discharged ammonia. The top trace is theconventional transmission spectrum showing two strong NH3 lines. The two middle traces demonstratethe increase in sensitivity gamed by using frequency modulation. With the discharge on a number ofnew lines appear in the spectrum. Three of these are due to NH2 as is shown in the lower trace, whichwas taken using Zeeman modulation.

among the (OOO), (lOO), and (001) vibrational states resemble each other closelyand even the values of the Coriolis parameter 2 are alike (2 = -0.303 cm- forNH:! and -0.319 cm- for Hz0 (22)). One can calculate from the known forcefiefd of Hz0 that the Coriolis coupling constant 6, is zero (22). A similar calculationfor NH*, using approximate force constants ffr = 6.2 md/& fe = 0.6971 mdA,fie = 0.224 md, frr = -0.100 md/A) gives a value of sfs = 0.00071 or & = 0.012cm-, well within the experimental range of 0.083 + 0.088 cm-.


11/14

110 AMANO, BERNATH, AND MCKELLARTABLE III

Molecular Parameters for NH2 (in cm-)(000) sta tea flOOl sta teb (001) atateb

2 3 . 6 9 3 4 412.95194

8 . 1 7 2 8 40 . 0 2 1 9 8 3

- 0 . 4 1 5 5 00 . 1 0 5 2 70 . 1 0 0 20 . 4 2 1 20 . 6 4 6 0

-0.113-0.620

0.411-0.309111-0.045173

0.40240 . 1 0 9 4

3219*331c:> 330?.270~1>23.12938179) 22.64616(%2312.76959(72) 12.878~4(8~)

%.01584(35) 8.05815147)0.020530(120) 0.033887~81~

-Q.440~(74) -~.4054(90~0.1007(15) 0.1080(1710.0801(37) O.lOQO(98)0.3921(33) 0.4126(121)0.942(61) 0.646OQ0.326157) -0.113C

-19.3(X> -0. 62QC0.89(28) O.lil?C

-0.29605(109) -0,28319(166~-0.04442(55> -~.0454~(117~-0.18(42) -0.25(72)

0.0599(102) 0.0963(141)Coriolis Parametersd0.083(88)

aGround state parameters from Birss et al. (27). Higher orderof (21)ara meters not shown her e were fixed at the valuesfor all thr ee states. Note tha t the definition ofA! used her e is opposite to tha t of (21). th e sign of

bPressnt results. Uncertainties in parentheses aredeviat ion from the least squa res fit, expressed inthe last quoted digit.one standardunits of

Pa ram eter fixed at its ground state value.dOnly he r e l a t i v e sign of the Coriolis parameters is significant.

The analysis by Flaud and Gamy-Peyret (22) of the vl and v3 bands of Hz0 alsoincluded the 2~~band. There is a substantial homogeneous (Fermi-type} interactionbetween the (020) and (100) states in I&O, as well as small higher-order Coriolis-type interactions between (020) and both (100) and (001). Interactions with (020)were not included in the present analysis and we did not notice any adverse effectsfrom this omission. Similarly, no evidence of perturbations was noted in the analysisof the (020) state of NH2 made by Birss et at. (5). One reason that the interactionswith the (020) state were not apparent is that the data were limited to fairly lowvalues of N and lu: (6 and 4 in the present work, and 8 and 5 in (5)). Furthermore,the effects of a homogeneous interaction are relatively easily absorbed into othermolecular parameters (this could account for the rather large change in HNK be-tween (~0) and (100) in Table III). The effect of the Fermi-type interaction


12/14

v, AND ~3 BANDS OF NH2 111between (020) and (100) has been noted for higher N by Vervloet and Merienne-Lafore (6) as a shift in the &, level of ( 100) perturbed by 826 of (020).

An interesting difference between NHz and Hz0 is that we observed the v, bandto be considerably stronger than v3, whereas just the opposite is true for HzO. Inorder to examine the relative band intensities quantitatively, we have measured the221- 212 ransition of vl and the l,, - 212 ransition of v3. These lines are especiallysuitable because they appear close together in the spectrum, they share the samelower level and they are Zeeman modulated with about the same efficiency. Fromtheir measured intensities, we estimate the ratio of the vibrational transition mo-ments to be about I~r/p~l = 4 for NH2, in contrast to the known value of Ip1/p3(= 0.223 for Hz0 (24). We have no explanation for this somewhat surprising result.

Table IV lists term values for NH1 rotational levels in the (000) (100) and(001) vibrational states calculated using the constants of Table III. Each excitedstate energy in Table IV is followed by a number which gives the percent mixingof its eigenfunction from the other vibrational state. That is, a value of 0.0 indicatesnegligible Coriolis mixing and 50.0 indicates complete mixing. Note, for example,

TABLE IVCalculated Energy Levels (in cm-) for NH,

0 0 0 3219.371 0.0 3301.110 0.0--i-Y--

0.000

:29*02t:63:5x.::8%120.744124.725wtg";154:615243.652244.066195.667197.561243.462262.165277.234330.407%':3t417:3en256.466287.266:::.:z393:0o7435.375443.666&2:%:634:951634.556

21.14332.00636.500

155.025;tt%:195.752197.676',g*s;z._277.59733l*O74132.541:::*:tt206.5701"53'~%i3as:oos:%z444:243826.e41527.119b36.277CJb.252393.295

0 1: 111:22' Y 22: :2 2 03 0 33 133:;:: : :3 3 0

:0 4: 4: 2:

a : 2: 314 a:

3 0 5: :: 2 4

3240.140 0.0 3240.174 0.0 3322.032 0.0 3322.066 0.03250.429 0.0 3250.651 0.0 3331.733 0.0 3331.944 0.03255.166 0.0 3255.421 0.0 3336.533 0.0 3336.779 0.03250.371 0.0 3260.427 0.0 3362.472 0.0 3362.530 0.03207.255 0.0 3207.395 0.0 3366.526 0.1 3360.964 0.13301.457 0.0 3301.655 0.0 3353.209 0.0 3353.403 0.03332.204 0.0 3332.714 0.0 3412.165 0.0 3412.675 0.03333.409 0.1 3333.997 0.1 3413.576 0.0 3414.061 0.03337.959zs E3394.5493400.4033457.7153457.064

3330.065

3458.644

8::X2823420.2543423.658:xi:: .3451.4323535.5153535.784

3420.3343423.776 i:;.i:;0:o.33411.4973413.214:tf:'taz:3491:1123542.0643543.981:t%%~ .

3411.5923413.327

3493.9213t:x:3550:1173073.7373620.2053621.6703702.0233702.0661,6-:,0:o4.3$:%0100.0

3500.6343%%3577:591

3tX3z:3649:lZJz'4:':3372b.274

3910.920

0.1X:X24.52:::50.922::ii::

3553.3013503.7733649.367z%'::o'3726:597x'e::-:::3510:5163912.13b3912.1423605.331Jb00.520xn3.g3.527:3793053.9613074.161p'o.~cl;4040:9214040.905

3605.314 0.03605.597 0.03690.202 0.03696.205 0.13743.036 0.33776.339 0.43791.478 5.73561.544 16.33567.905 20.03972.138 4.43972.145 4.1 76S.310

8 : x6 9 2.0IX JNOTE: The numbers following the energies are percent mixings with the other vibratonal state (see text).


13/14

112 AMANO, BERNATH, AND MCKELLARthat the 542 evel of (100) is virtually completely mixed with 532of (OOl). The(100) state levels in Table IV agree well with those of Vervloct and Merienne-Lafore (6) derived from the electronic spectrum. Moreover, the perturbed levelsof (100) and (001) identified by those authors are all confirmed by our results,including the (100) 542* (001) 532pair mentioned above.While recording the spectrum of NH*, transitions belonging to the fundamentalband of the NH radical were also observed, these results will be reported separately(25). An improved version of the laser system and Zeeman-m~ulation cell havesince been used to study CH3 (26), SH (27), and SO (28). These results, and thosereported here, illustrate the utility of the difference frequency laser system combinedwith a Zeeman-modulated discharge cell for the study of free radicals in the in-frared.

ACKNOWLEDGMENTSWe would like to thank T. Oka and A. Karabonik for the construction of the difference frequencylaser system, and P. Brechignac for his help in the design and construction of the Zeeman-modulationcell. We are also grateful to M. Vervfoet for ~mmunication of the results of Ref. (6) prior to publicationand to J. W, C. Johns and M. Vervloet for their ~mmen~ on the manuscript.

RECEIVED: February 2, 1982

Strictly speaking, the Fr components of these two levels should be interchange in Table IV, sincetheir mixing coefficients exceed 50%. However, we have adopted a more natural labeling that conservesa reasonable spin splitting.

REFERENCES1. K. DRE ~SLER ND D. A. RAMSAY,Philos. Trans. Roy. Sac. London Ser. A. 251,553-602 ( 1959).2. J. W. C. J O H N S ,D. A. R AMSAY, AND S. C. Ross, Can J. Phyr. 54, 1804-1814 (1976).J. G. W. HILL S, R. S. Low + J . M. COOK, AND R. F. CURL, JR., J. Chem. Pkys. G&4073-4076(1978); G. W. HILLS, J. M. COOK, R. F. CUR L, J R., AND F . K . IIT~E L, . Ckem. Pkys. 65,823-

828 (1976).4. M. KROLL,J . Ckem. Phys. 63, 319-325 (1975).5. F. W. BIRSS,M.-F. MERIENNE-LABORSD. A. RAMSAY, AND M. VER VLOE T,I . Moi . Spe cr r osc .8!$493-49s (1981).

6. M. VERVLOET ND M. F . MER IEN NE-LAFORE ,an J. Phys. 49,495s (1982).7. D. E. MILLIOANAND M. E. J ACOX,J . Ckem. Pkys. 43,4487-4493 (1965).8. K. KAWAGUCHI,C. YAMADA, E. HIR OTA, J . M. BRO WN, J . BUIT ENSH AW,C. R. PARENT,AND

T. J . SEARS,J . Mol. Sp ct r osc . 81, 60-72; G. W. H ILL S AND A. R. W. M CKE LLAR ,J. Mol.Spectrosc. 74,224-227 (1979).9. A. S. P INE, f. Opt. Sac. Amer. 64, 1683-1690 (1974); 66,97-108 (1976).IO. P. BERNARDAN D T. OKA, J. Mel, Spectrwc. 75, 181-196 (1979); A. R. W. MCKELLARAND T.OKA, Can J. Pkys. 56, 1315-1320 (1978).Il. T. OKA, Pkys. Rev. Left. 45,531-534 (1980).I.7. P. BERNATHAND T. AMANO, Pkys. Rev. Left. 48, 20-22 ($982).13. M. WONG, P. BERNATH,AND T. AMANO, J. Ckem. Pkys., in press.14. E. HIROTA, n Chemical and Biochemical Applications of Lasers (C. B. Moore, Ed.), Vol. 5, pp,39-93, Academic Press, New York, 1980.


14/14

Y, AND vs BANDS OF N H2 11315. G. LITFIN, C. R. POLLOCK, R. F. CURL, JR., AND F. K. TITTEL, J. Chem. Phys. 72, 6602-6605

(1980); J . PFEIFFER,D. KIRSTEN, P. KALKERT, AND W. URBAN, Appl. Phys. B 26, 173-177(1981).

16. A. S. PI NE , M.I.T. Lincoln La bora tor y Repor t No. NSF /ASRA/DAR-78-24562 (1980).17. C. AMIOT AND G. GUELACHVILI,J. Mol. Spectrosc. 59, 171-190 (1976); C. AMIOT, J. Mol.Spectrosc. 59, 191-208 (1976).

18. A. R. W. MCKELLAR, J. Chem. Phys. 71,81-88 (1979); R. S. LOWE AND A. R. W. MCKELLAR,J. Chem. Ph ys. 74, 2686-2697 (1981).

19. J. K. G. WATSON, in Vibra tional Spectra an d Str uctur e (J. R. Dur ig, Ed.), Vol. 6, pp. l-89,Dekker , New York, 1977.

20. J. M. BROWN AND T. J. SEARS,J. Mol. Spectrosc. 75, 11 l-133 (1979).21. F. W. BIRSS,D. A. RAMSAY, S. C. Ross, AND C. ZAULI, J. Mol. Spectrosc. 78, 344-346 (1979).22. J. M. FLAUD AND C. CAMY-PEYRET, J. Mol. Spectrosc. 51, 142-150 (1974).23. T. TANAKA AN D Y. MORINO, J. Mol. Spect rosc. 33, 538-551 (1970).24. J. M. FLAUD AND C. CAMY-PEYRET,J. Mol. Spectrosc. 55, 278-310 (1975).25. P. BERNATH AND T. AMANO, in preparation.26. T. AMANO, P. BE RNATH, C. YAMADA, Y. EN DO, AND E. HIROTA, in preparation.27. M. WONG, T. AMANO, AND P. BERNATH, n preparation.28. M. WONG, T. AMANO, AND P. BERNATH,J. Chem. Phys., in press.

Date post:	06-Apr-2018
Category:	Documents
Upload:	hojhiema
View:	215 times
Download:	0 times

T. Amano, P.F. Bernath and A.R.W. McKellar- Direct Observation of the nu-1 and nu-3 Fundamental...

Documents