1170 J. Opt. Soc. Am. B/Vol. 3, No. 9/September 1986

Fourier-transform spectroscopy of NH: the cII1-a'Atransition

R. S. Ram and P. F. Bernath

Department of Chemistry, University of Arizona, Tucson, Arizona 85721

Received February 18, 1986; accepted May 28, 1986

The high-resolution Fourier-transform emission spectrum of the c 1 II.a 'A transition of NH was recorded andanalyzed. Improved line positions and molecular constants were determined from the 1-0, 0-0, and 0-1 vibrationalbands. In addition to the rotational constants, lambda-doubling parameters for both the c 11 and a 'A states wereextracted from the line positions.

INTRODUCTION

The NH molecule is, along with CH and OH, one of the moststudied free radicals. The major reason for this interest isthe wide variety of environments-from flames' to stellaratmospheres 2 -in which NH can be found. During our re-cent reanalysis of the A31-X 32:- transition of NH,3 we alsodetected the c'I-a 'A system of NH. In this paper wereport on the 0-0, 0-1, and 1-0 vibrational bands of the c-atransition. The high-quality Fourier-transform data pro-vide some improvement in molecular constants, includingextraction of the lambda doubling in the a 'A state.

The NH molecule was first observed by Eder4 in 1893through the A 311-X 32- transition near 3360 A. The c'l1-a 'A transition near 3240 A was first observed in the 1930'sby three groups at about the same time: Nakamura andShidei,5 Dieke and Blue,6 and Pearse.7 Hanson et al.8 havestudied the corresponding ND spectrum, as have Cheung etal.9 More recently, Ramsay and Sarre'0 have analyzed the0-1 band. The hyperfine and lambda-doubling parametersof the c 'H-a 'A were observed in a laser experiment on amolecular beam by Ubachs et al.11

A number of transitions connecting to either c 11I or a 'A,such as dl+-c 1 (Refs. 12 and 13) and cII-b3 +,l4 havebeen observed. The singlet and triplet manifolds were firstconnected through detection of the bZ:+-X 3

1- transitionby Masanet et al.15 The most accurate measurements ofthis transition were made at high resolution by Cossart.16

The combination of Cossart's lines' 6 with data from Ref. 14and the c-a lines obtained in this work result in an a 'A (v =0, J = 2)-X3 3- (v = 0, J = 1, N = 0) separation of12 688.38(10) cm-'.

The direct a 'A-X 32; transition has also been found.' 7

Hall et al.'8 detected the fundamental infrared vibration-rotation transition of the long-lived aA state. RecentlyLeopold et al.' 9 observed the far-infrared laser magneticresonance spectrum of NH and ND in the metastable a 'Astate.

The NH molecule is produced by electron bombardmentor photodissociation of a wide variety of molecules, includ-ing NH3, N2H4, CH3NH2, NH3, and HNCO.20-23 These ex-periments allow the lifetimes 2 ,22 of excited states of NH tobe determined and provide, in addition, fundamental infor-

mation about electron and photon dissociation processes.For instance, Alberti and Douglas23 found that vacuum-ultravolet photodissociation of NH3 produced an anomalouspopulation distribution in the NH product. By monitoringthe c 1II-a 'A fluorescence, they found that the two lambda-doubling components of the c 1II state are unequally popu-lated.

EXPERIMENT

The experimental details are provided in our paper on A 3II-X3y-.3 Briefly, the c 11-a 'A emission of NH was excited ina water-cooled copper hollow-cathode discharge lamp oper-ated at 440-mA current. A continuous flow of 4.5 Torr ofhelium, 0.12 Torr of nitrogen, and 0.04 Torr of hydrogenprovided a strong NH signal. The molecular emission wasrecorded with the Fourier-transform spectrometer associat-ed with the McMath solar telescope of the National SolarObservatory24 at Kitt Peak. The spectrum had an unapo-dized resolution of 0.042 cm-1 and was calibrated by usinghelium atomic lines present in the discharge.

RESULTS

The spectral region 24 500 to 33 500 cm-' has structure thatis due to two electronic transitions of NH, namely, A3HIIX3

7- and c1-alA. The structure of the bands of theA 311,-X 3 2- transition is strong compared with the structureof the c-a transition and appears with a signal-to-noise ratioof >1000. The structure of the 0-0, 1-0, and 0-1 bands ofthe c-a system was easily identified owing to the characteris-tic splitting of each line into two lambda-doubling compo-nents. A part of the spectrum of the 0-0 band of this system(near the R head) is given in Fig. 1. The 0-0 band of thissystem is about five times stronger than the 0-1 band andten times stronger than the 1-0 band. The 1-1 band is alsopresent, but most of the rotational lines are overlapped bythe strong 0-0 band of the A-X system. Therefore thisband was not included in the present study. The spectrumwas measured by using a computer program called DECOMPdeveloped at the National Solar Observatory. All lines werefitted by a nonlinear least-squares procedure to Voigt line-

0740-3224/86/091170-05$02.00 © 1986 Optical Society of America

R. S. Ram and P. F. Bernath

R. S. Ram and P. F. Bernath Vol. 3, No. 9/September 1986/J. Opt. Soc. Am. B 1171

R(7) R(6)I

za)

0

a)

0

440

a

--

C)

I

L

4-

'U

.0'U

30835.00 30845.00 30855.00 cm'

Fig. 1. A portion of the emission spectrum of the c I-a 'A 0-0band. Lambda doubling splits each line into two components (eand f). The band head occurs near 30 855 cm' at R(5).

shape functions. The estimated precision of measurementsfor strong, unblended lines of the 0-0 band is of the order of+0.001 cm-1, consistent with a line width of about 0.20 cm-'and a signal-to-noise ratio of about 200. For the weaker 1-0and 0-1 bands, which are frequently obscured by the spec-trum of N2 + lying in the same spectral region, the precisionof measurements is estimated to be ±0.005 cm-'. The abso-lute accuracy of the overall frequency calibration 3 is estimat-ed to be 0.003 cm-'. Our line-position measurements aresystematically 0.0051(31) cm-' higher than those of Ubachset al.11 Their absolute calibration using iodine may beslightly more secure than ours, which is based on a singlehelium line.3

We have observed the rotational structure up to J' = 18and J' = 11 in the bands involving v = O and v = of the c 'IIstate. The study of the d 2+-c ']I transition of this mole-cule by Graham and Lew'2 clearly established that the c IIstate is predissociated beyond J = 22 and J = 15 in the v = 0and v = 1 levels of the c If state. We were unable to observethe structure up to the point of strong predissociation be-cause of an inadequate signal-to-noise ratio. However, theintensity of the lines decreased much faster than predictedon the basis of a Boltzmann population distribution. Abroadening was observed for the lines involving higher Js.For example, the widths of the Qef(3 ), Qet(lO), and Q(18)lines of the 0-0 band are 0.194,0.198 and 0.243, respectively,and those of the 1-0 band for the Qf(4), Qf(7), and Qef(9 )lines are 0.240, 0.255 and 0.268 cm-1, respectively. Thelevels affected by predissociation2 2 show the broadeningmost noticeably. Application of the uncertainty principleto the state with the shortest observed lifetime of 41.7 nsec(Ref. 22) for J' = 9, v = 1 of the c ]I state predicts a broaden-ing of only 4 MHz (0.00013 cm-'), much too small to accountfor the observed increase in broadening.

A plausible explanation is that the high J' (and v' = 1)levels, which have emission intensities reduced by predisso-ciation, are highly pressure sensitive. A similar effect wasobserved for the A 3 Il-X 32- transition.3

The wave numbers of the observed rotational lines of the0-0, 1-0, and 0-1 bands are given in Tables 1, 2, and 3,respectively, where the branches are labeled according to thee and f parities25 of the initial and final states.

csI I

en o'

i: n 00

enl en

Lo Ico a CoI I I I I1-44 I

e CnD t- t- e e e -4 oo

0 t 'D M Lo M M CSq M Cen n e -4 0 t- t- >C en e eL 'den

O ' o~ en~ en ot ci1 -oe t- t C l C; - en.1CtoUjC eUI c ci-4 0 en e' -en

O LO "It I 0 0 0 0 t toCe D C C C C n e C i C C C

- L_ M -_ ~o a- C L -4 ICcS 'i' 0300O 0 3 - en ci cs en 03 C 0303 CO 0 ,0 C, D 03 M C0S0 '0 -

enr 0: r en en en C r ot- en en en cioo ' 03 1 0 en eD cs en L- L- t- t>en CInD C C) Cen D CS CS -0 e M 0 Cn

eo e e CO n en O e Cn en D C CS CS CI

L_ _ll

I I

LO C11

en 00en csen 0

n4 cor- c-00 CenD en

M 10 cS

O! CI V

V co -4

ci cq ai

~-'0O I

I

_) _

M t- Ien 0.

( ciaM MCS Cq

$1I I I ' - cs I c - I I I -

4000 en Lo en en Cq 03 t s 0 m0 c C 03 en -40 O c- LO O na4 ) C LO t- ~ C r- : r_ LO * I'dO o LO t- 't 0 C t- 00 t- CD -C4

0> 00 0 C 0 t t 00 00 C 00 3o o o en e e e e en en

CD C CID c o co co c o c c oM c c

cs I a - o r- c e ;e

Lo 0 , e o -40 c C C-4c c -4 en en oo eo 0 o 0 c

03 Ci C e 00 e CD O 00 0C-m ci 0o enjo oi e X o. s en c 000 t 00 t 000 0 0 r " o n c e en e n ce oo en e e e enr- _ L_ 'I CD CD LO U t M 0 "0:C

o ) o > o o > o > o C> o > o oo>

I

0 CLO enL_ COR'-4 t-L 00en enCD C>

LO"v I

rz cstz- ene 0

e b00 3

CS en

Lo : I I I S TFizPTjo~ t~ ~'05 ;v en en ne 4 n5 ~'t = UI O tO C=> C M CO - to CS

" IV t t-e 0 en 'I t -4 en 00

Lq I O~ aO L O OO~ Cl Ci C O~ Ilen en en en en en C: L- L- en en C 0 i0 o0000000000000000 C C C C C C C Cen en en en en en en en en en en enD en en en en

Lt- O '0 M 3 - 0 0 M en ci0O en cI jt - I I 0 C n I C Ir i e n r o 0 3 t 0 o 0 n 0 e n o c

0D C C> 11: -zv O to ML 00 -4 M L0 e CD 0 0 t- LO 'I 0 0M e 0 0

r M jo r o o e o o en en 03o L 00000 ei00 00 s>00 0 00 C en en en en en en en en en en en en en enC7 C C 0: C C C CD C C C

en-

C UD

_,RO

0:1 LO

LO OtC> CD

en c

c e M t e u nC rW e 0 o - cs e o ' en CD e 0 3 -4 -4-4--4-4 -4 -4-4-4 -4

NH: cn - aA (O, O)

R(8)

R(3)

5)

-o

0)

a0J

00* u

0

0o

0

a

a

.)

.04

on~

5

.0

oa a

+) 5)

0) aC C

I I

I I

1172 J. Opt. Soc. Am. B/Vol. 3, No. 9/September 1986 R. S. Ram and P. F. Bernath

Ni 4 CDCSc _I I --cI I

o o oo cm rm M CoCc ci to c - ci c

o R i' c O cri :ciO LO CS C i- t-

C c c c c co LOC S QC C SC

IC I CO I rz- ci oz c LO4 t Co c C 0 0 cD M co c 0Ci C 0 Cc O CC

0 OM cCD c e-4i- C) o o oo a) -t-* - cD CO CD ccI ci ci cS c

II I

0 tr o c c c _ c cc ci ci -c_oo I Ncs cc I o- cc H cc r- cc C 0 C w o UI Lo L D>

c c ,- CD r 1 c cc cc o c c s0 cc l cc c oci cc 0 i C ci c cccc t 9 cc cc cs ci -44 o~ cc cc oo ccL- t- - -r r c cc co C cOcs .s cs .s cT . cs c c c c c c c

-- __ Occ. -40C

t- c ci'-4 t- 0)cc cc M

I:- 5- 5:-ci ci ci

0 ccO 00c 0cc o00 >

00LO I

C- -cli ci1

CO co t- cc c

-- C9-4 1 I I I I I I I i

,,I, CD > - U')M -1 '. "dM00t- 0 ot cco 5:- r) b oi -4 cs tc c c c 0 B t -c o 9 cc O ' c c~i 0 i ci cc cc t- L- 0 c ci ksi,,I: cc~ (c t- C) 0l~ :: '4l -! ccR 0' c c t-- " ci c M M ' ci cc

M M "W CO Co CS -411 w c c c ci i '-4 -4 ) M-L- r- t- r- - t- r- r- r- co cc

C C C C C C c cS Ci ci CS CS

_ CIOG RQ LO _ _ _5:-r:cis -.4I I csiccI I cnC Dc I 00 IM) - -4 cc c c c c c - s -t- cc S ~ c

H~~~~~~( oow>rto osc CD > C cc cco co cO 0) co cc o- 40 cc 0 0 0 ci cc tcc 0 0 M4 cc - U- 0i M L W' 'c± ci cc cc

c c c c 0 ) - I- -4 r- c c 0 ) 0 c c c ) c i

cD C c . . c c. i . . .i co c.

0-4-4 -4l. LI I OI C9

oo cc 0 00cq cc0 t- Cc 0

c t- oo r M

tc Cc c cc cc c cci r- t- c c ci tM

CS Cli CS Cq CS C9

CD O 0D 0 0 to

_0 __

cc CD --t 'I I ci occo Z r Izq 00 5 S C U IQ co

CO - cc c 00) ' -cs- cc i i cc cc L-cc3 -! cc~ ccc cc~ cc cc0) cc r- cc C, '.- cc '4

to cc cc cc cc cc cc ccL-- - L- t- 5:- t-c-- C -* . c ci ci ci cq c

I I

cccc oo C r _ - Ccc-4-- q -4 -4 cc

H 1 I C -c 1 Ci I II I ci C I

) cc cc ' -4 C- c 00 cc c CD Cc 0 cS 'cc -4 c- cc cc cc Cc co t- 00

cc cc o oO P ar c cotU'4 cc c- cc ) r 0i t ) cc i r ccI t- cc 0lll CD r- oo M M aM oo t- to 11: - 00 et4P 0CD CD w to CD -C to to- C CO Co C' L- ci ci ci ci . . . . ci c i . c c ci

C iS c c cc c 0 ) 0 -4 C c'll cc c C- 00-4 - -4 - -4 -4 -4

z

-

02

av ~ ~~~~~~a

140 _

0

o ;~~~~~~~c

C.) 0~~~~~~C

_) -

E4-4

0) 0

.

Qe Q~~~~~~ao

n ~~~~~~~~a)o <E

0 a)4) =4

,:

C)

4) a)

.04.

4, gQ~4)0a

- c'' cc-

to -cc c o cc0 to: "I M C) I O

0 cc cc c a" cc cc c cMLq cs o R oo oo s Cc C cc cc cc cc -0 .c 0 . . . . csoo M M c c M cocs ci cs ci cs ci cs cscc cc cc cc cc cc cc cc

I

cc ~5cc ccco -41-40C'-40Ccs cicc Cc

°- ci °cc_ _ -o c - _II Icct-Il--w -

0 er O O CD U cc O Cc cco i- - cc -4 c 0 c c cc

0 cc 0 cs t oo e i lCi -40cvcc i- o- cc cc O cc ci -4cs ci c c cs cs ci cs ci ci cscc cc cc cc cc cc cc cc cc cc cc

cct- LtoL 11ci11 (

-CD 4 -4---4-4 0°

'Cl .o- _ ___ 5Uc -4 ccO - - cc c ccc-4 -14 C-4c w O c Icc Ci Ci q c 00 c c I O Cc c O c t I cc cc

cc c c0 - -cc cc cc cc Cc C cq cq C C C C c CiCI c cc cc c c c C c cc

cicc

cdob-I -cc 00 cc cc0 0' cc

o C C o CO4 -4 CO oM CO o

-I - --t c - cI

cc-4llLqCcccLo cDoR 0O t Ilco Lo _Co V _r- cCc CO C t- c - 4 ccc cc cc c- z- o C cc ccci ci ci c ci ci ci ci . C icc cc cc cc cc cc cc cc cc oo

o CI c cc CS

CCS CSC fCSC SC

cc, o= c cc cc 0

Ico I --o a)c I Icct ci. 0 1-1 cc o cc

zcq o~ c cc r- c-ci ci ci ci ci c cicc cc cc cc Cc Cc Cc

- .0M cccc 0ci ci

CID ccM MCS CS1co 0)

",-C, 0 to c c

0) c- ci cc cc5 c ci00 U: CD q to 7c cc c c c c o c

I cc 0 ) 0 ci cc cc

cc c i c i cc M ) M 0

0 ) . . c c c c C' C-ci ci ci c ci ci ccc cc cc cc cc cc cc

C i c c - c c C c C -c c 0 ) , - c i

17w

z0a)0

-I

L

'4-40

-4

0

qdM

Ia)

..'4-40

UW,M

a)

0

a)

44

4

CD

Ev

'4~0za)

C.)

~0

4)

4,

0R

4)

.5

0

0

0.

'0

-4)

)

4.I

.0 0

E-O

Vol. 3, No. 9/September 1986/J. Opt. Soc. Am. B 1173

ANALYSIS AND DISCUSSION

The rotational constants for the c lII and a 'A states weredetermined from a nonlinear least-squares fit of the data inTables 1-3. All three vibrational bands were simultaneous-ly fitted. The rotational energy expressions for these twostates are defined as follows:

(for the '1II state)

F,(J) = B,[J(J + 1) - 1] - D,[J(J + 1) - 1]2

+ H[J(J + 1)- 1]3 + L[J(J + 1)-14

4 12q, + qDJ(J + 1) + qH,[J(J + 1]2IJ(J + 1),

(for the 'A state)

F,(J) = BV[J(J + 1) - 41 - D,[J(J + 1) - 4]2

+ H,[J(J + 1) -4]3 + L,[J(J + 1) -4]4

i /2q,[J(J + 1)]2.

In these equations the upper (lower) signs correspond tothe e (f) parity 25 components and the constants B, D, H, andL have their usual spectroscopic meaning. The lines of thefundamental vibration-rotation band of the a 'A state re-ported by Hall et al.'8 were included in our fits so we couldderive more-accurate spectroscopic constants. The data ofHall et al.18 included R-branch lines up to J" = 34 but onlytwo Q-branch lines and no P-branch lines. The rotationalconstants obtained in the final fit are given in Tables 4 and 5.

It was found that the lambda-doubling constants, q, of thetwo vibrational levels of the a 'A state have appreciable mag-nitude. Lambda-doubling parameters for 'A states are rare-ly determined because they are so small. Recently Ubachset al." found q to be -4.82(31) kHz, in excellent agreement

Table 4. Rotational Constants (in cm-I) for the a'AState of NH

Constantsa v = V = 1

TVb 0.0 3182.7879(16)B, 16.432551(76) 15.814214(87)

104 X Dv 16.7309(57) 16.4156(69)108 X Hv 11.54(22) 11.01(24)

1012 X Lv 6.0(34) 3.8(30)107 X q -1.62(23) -1.46(21)

a Values in parentheses represent one standard deviation in the last digits.b The singlet-triplet splitting is 12 688.39(10) for a 1A (v = 0, J = 2)-X32-

(v = 0, J = 1, N = 0) obtained from the line positions of Tables 1-3 and Refs.14 and 15.

Table 5. Rotational Constants (in cm-') for the c'1HState of NH

Constantsa v = 0 V = 1

Tub 30704.0818(16) 32825.5375(67)B, 14.15158(10) 12.86063(66)

104 X D, 22.462(11) 27.22(19)108 X Hv -5.89(47) -52.5(20)109 x L, -0.2814(73) -2.02(71)103 X qv 16.711(38) 17.67(23)106 X qDv -3.92(38) -14.0(53)108 X qHv 0.216(88) 9.1(29)

a Values in parentheses represent one standard deviation in the last digits.b Measured relative to a 'A (v = 0).

with our value of -4.86(69) kHz for v = 0 of a 'A. There isalso good agreement between our lambda-doubling parame-ters and those of Ubachs et al." (except for qH) for the c 1IIstate. However, our rotational constants differ somewhatfrom those of Ubachs et al." For example, our Bo is14.15158(10) for the c '1 state, while Ubachs et al. obtained14.14203(22). Our Bo for the a 'A state is in excellent agree-ment with the more accurate value determined by Leopoldet al.'9 in a far-infrared laser magnetic resonance experi-ment.

Our value of AG",1 2 for a 'A is 3182.7879(16), comparedwith the value of Ramsay and Sarre 0 of 3182.758(35) cm-'and with that of Hall et al.18 of 3182.7768(37). For AG'1/2our value is 2121.4557(69), compared with that of Grahamand Lew,12 2122.54 cm-'.

CONCLUSION

The 0-0, 1-0, and 0-1 bands of the c 11-a 'A transition ofNH were observed by high-resolution Fourier-transformemission spectroscopy. Improved molecular constants wereextracted, including the determination of the lambda-dou-bling parameters for v = 0 and v = 1 levels of the a 'A state.

ACKNOWLEDGMENTS

We are grateful for the expert technical assistance of RobHubbard in acquiring the NH spectrum. We thank JimBrault for the helium spectra employed to calibrate our dataand his interest in this work. This work was supported byfunding from the U.S. Office of Naval Research.

REFERENCES AND NOTES

1. W. R. Anderson, L. J. Decker, and A. J. Kotlar, Combust. Flame48, 179 (1982); M. S. Chou, A. M. Dean, and D. Stern, J. Chem.Phys. 76, 5334 (1982); J. A. Miller, M. C. Branch, and R. J. Kee,Combust. Flame 43, 81 (1981).

2. J. L. Schmitt, Pub. Astron. Soc. Pac. 81,657 (1969); R. W. Shaw,Astrophys. J. 83, 225 (1936); D. L. Lambert, J. A. Brown, K. A.Hinkle, and H. R. Johnson, Astrophys. J. 284, 223 (1984).

3. C. R. Brazier, R. S. Ram, and P. F. Bernath, J. Mol. Spectrosc.(to be published).

4. J. M. Eder, Denkschr. Wien Akad. 60, 1 (1893).5. G. Nakamura and T. Shidei, Jpn. J. Phys. 10, 5 (1935).6. G. H. Dieke and R. W. Blue, Phys. Rev. 45, 395 (1934).7. R. W. B. Pearse, Proc. R. Soc. London Ser. A 143, 112 (1933).8. H. Hanson, I. Kopp, M. Kronekvist, and N. Aslund, Ark. Fys. 30,

1 (1964).9. W. Y. Cheung, B. Gelernt, and T. Carrington, Chem. Phys. Lett.

66, 287 (1979).10. D. A. Ramsay and P. J. Sarre, J. Mol. Spectrosc. 93, 445 (1982).11. W. Ubachs, G. Meyer, J. J. ter Meulen, and A. Dymanus, J. Mol.

Spectrosc. 115, 88 (1986).12. W. R. M. Graham and H. Lew, Can. J. Phys. 56,85 (1978).13. R. W. Lunt, R. W. B. Pearse and E. C. W. Smith, Proc. R. Soc.

London Ser. A 155,173 (1936); F. L. Whittaker, Proc. Phys. Soc.London 90, 535 (1967); G. Krishnamurty and N. A. Narasim-ham, J. Mol. Spectrosc. 29, 410 (1969).

14. R. W. Lunt, R. W. B. Pearse and E. C. W. Smith, Proc. R. Soc.London Ser. A 151,602 (1935); F. L. Whittaker, J. Phys. B 1,977(1968).

15. J. Masanet, A. Gilles, and C. Vermeil, J. Photochem. 3, 417(1974); Chem. Phys. Lett. 25, 346 (1974).

16. D. Cossart, J. Chem. Phys. (Paris) 76, 1045 (1979).17. F. Rohrer and F. Stuhl, Chem. Phys. Lett. 111, 234 (1984).

R. S. Ram and P. F. Bernath

1174 J. Opt. Soc. Am. B/Vol. 3, No. 9/September 1986

18. J. L. Hall, H. Adams, J. V. V. Kasper, R. F. Curl, and F. K.Tittel, J. Opt. Soc. Am. B 2, 781 (1985).

19. K. R. Leopold, K. M. Evenson, and J. M. Brown, J. Chem. Phys.85, 324 (1986).

20. U. Blumenstein, F. Rohrer, and F. Stuhl, Chem. Phys. Lett. 107,347 (1984); A. Hofzumahaus and F. Stuhl, J. Chem. Phys. 82,3152, 5519 (1985); H. K. Haak and F. Stuhl, J. Phys. Chem. 88,2201, 3627 (1984).

21. I. Tokue and Y. Ito, Chem. Phys. 52, 47 (1980); 79, 383 (1983);89, 51 (1984); J. Phys. Chem. 88, 6250 (1984).

R. S. Ram and P. F. Bernath

22. W. H. Smith, J. Brzozowski, and P. Erman, J. Chem. Phys. 64,4628 (1976).

23. F. Alberti and A. E. Douglas, Chem. Phys. 34, 399 (1978).24. The National Solar Observatory is operated by the Association

of Universities for Research in Astronomy, Inc., under contractwith the National Science Foundation.

25. J. M. Brown, J. T. Hougen, K. P. Huber, J. W. C. Johns, I. Kopp,H. Lefebvre-Brion, A. J. Merer, D. A. Ramsay, J. Rostas, and R.N. Zare, J. Mol. Spectrosc. 55, 500 (1975).