Nuclear hyperfine interactions in the microwave spectrum of aluminium isocyanide

24 October 1997

ELSEVIER

CHEMICAL PHYSICS LETTERS Chemical Physics Letters 278 (1997) 9-15

Nuclear hyperfine interactions in the microwave spectrum of aluminium isocyanide Kaley A. Walker, Michael C.L. Gerry Department of Chemist~', The Unit'ersi O' of British Columbia, 2036 Main Mall, Vancout'er, BC, V6T IZI Canada

Received 8 August 1997

Abstract The pure rotational spectrum of A1NC has been measured between 11.9 and 24.0 GHz using a pulsed jet Fourier transform microwave spectrometer. The molecule was produced using a laser ablation source, in which the ablated A1 metal reacted with cyanogen present in the Ar backing gas of the jet. Hyperfine structure due to both the 27A1and 14N nuclei has been observed. Nuclear quadrupole coupling constants and nuclear spin-rotation constants have been determined for both nuclei along with an AI-N nuclear spin-spin coupling constant. The electronic structure of AINC has been investigated by estimating the degree of sp-hybridisation of the AI and N bonding orbitals. Comparisons have been made to similar linear metal isocyanide and aluminium halide species. © 1997 Elsevier Science B.V.

1. Introduction The spectra of refractory-element molecules in circumstellar clouds provide information which will further our understanding of the chemistry of these elements in space [1]. Laboratory spectra provide search parameters for new species and can be used to identify (and verify) lines in the spectra of interstellar sources. The detection of NaCN [2] and MgNC [3,4] in the circumstellar envelope of IRC + 10216 has raised the possibility that other metal cyanides and isocyanides might be present in interstellar sources. Specifically, A 1 N C / A 1 C N has been suggested [5]. This seems promising because A1 and Mg have comparable cosmic abundances, and AIC1 [6] and AIF [6,7] have been detected in IRC + 10216. AINC and AICN were first detected by mass spectroscopy by Gingerich in 1967 [8]. Subsequently, three ab initio studies have been made of the

structures of these molecules [5,9,10]. These studies predict that AINC and AICN should be linear, and that AINC should be more stable than A1CN by about ~ 2 3 k J / m o l . Two experimental studies have been reported quite recently. Robinson et al. have measured the millimeter wave spectrum of A1NC in both the ground vibrational state and excited states of the bending mode, v 2 [11]. Fukushima has reported investigation of the 1A5-5(1£+ electronic transition of A1NC and A1CN by laser induced fluorescence [12]. In this article, we report the measurement of the rotational spectrum of AINC in the frequency range 11.9 24.0 GHz. The instrument used was a pulsed jet cavity Fourier transform microwave ( F T M W ) spectrometer. The samples were prepared using our new laser ablation system [13] with the nozzle mounted in the cavity mirror. The first measurements of the hyperfine structure in A1NC due to the 27A1

0009-2614/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved. Pll S0009-2614(97)00951-2

10

K.A. Walker, M.C.L. Gerry/Chemical Physics Letters 278 (1997) 9-15

and 1 4 N nuclei have been made, and nuclear quadrupole coupling, nuclear spin rotation, and nuclear spin-spin constants have been determined.

2. Experimental details The experiments were performed using a BalleFlygare type Fourier transform microwave (FTMW) spectrometer [14] which has been described in detail elsewhere [ 15]. In brief, the spectrometer consists of a cavity formed by two spherical aluminium mirrors 24 cm in diameter separated by approximately 30 cm. One of the mirrors is held fixed while the other is movable, so the cavity can be tuned into resonance with the microwaves at the excitation frequency. The pulsed jet is injected into the cavity parallel to the axis of microwave propagation. This "parallel" configuration increases the sensitivity of the spectrometer [16], and, since the line widths obtained by this technique are very narrow ( ~ 8 kHz), each transition appears as a doublet due to the Doppler effect. A1NC was produced using our new laser ablation system with the nozzle mounted in the fixed mirror, which has been described in detail elsewhere [13]. For this series of measurements, the second harmonic of the Nd:YAG laser was used to ablate an AI rod (Goodfellow 99.999%) held in a stainless steel nozzle cap 5 mm from the orifice of a General Valve Series 9 pulsed nozzle. A schematic diagram of the nozzle cap is given in Ref. [13]. The nozzle was operated with a backing pressure of 4 - 6 atm. A1NC was produced by reacting the ablated metal with cyanogen present as 0.1% in Ar carrier gas. The signals were so strong that the J = 1-0 transitions were easily observable in a few measurement cycles; 10 averaging cycles produced a power spectrum with 50:1 S / N . Transition frequencies of unblended lines were determined by averaging the line positions of the Doppler components obtained from the power spectrum. Frequencies of closely spaced or overlapped lines were obtained by fitting to the time domain signals [17]. Measurement accuracy is estimated to be better than _+ 1 kHz. The microwave synthesiser is referenced to a Loran frequency standard, which is accurate to 1 part in 1012.

3. Observed spectra and analyses When this study was undertaken no experimental results had been presented, so the transition frequencies were predicted using results from the theoretical study by Ma et al.[5]. Experience with the similar species MgNC provided some added insight into how the theoretical results could be employed. For MgNC, it was found that calculations made at the S D C I / T Z 2 P level of theory produced a rotational constant B that was about 27 MHz lower in frequency than that obtained from experiment [18]. In their theoretical study of A1NC, Ma et al. found that B 0 and Be differed by only about 5 MHz. Consequently the Be constant predicted for A1NC using the same level of theory as for MgNC was chosen to calculate the initial search range. To estimate the magnitude of the hyperfine splitting, eQq(27A1)from A1F [19] and eQq(14N) from MgNC [20] were included in the prediction. The J = I - 0 transitions were found less than 40 MHz from the predicted values. Two rotational transitions, J = 1-0 and J = 2-1, were available in the frequency range of our spectrometer. 27 lines were measured for the main isotopomer 27AlI4N12C (98.53% natural abundance). Species containing other isotopes of N and C were not sought. Nuclear hyperfine structure due to both 27A1 ( I = 5 / 2 ) and 14N ( I = 1) was observed. Since the splittings due to 27A1 w e r e larger than those due to 14N, the lines were easily assigned in terms of the coupling scheme J + Ial = F1; F1 + IN = F. An overview spectrum of the J = 1-0 transition, shown in Fig. 1, shows the relative magnitudes of these splittings: the 27A1 quadrupole coupling produces three lines which are then split further by the 14N quadrupole coupling. An expanded view of the FI = 7 / 2 - 5 / 2 section of the J = 1-0 transition is given in Fig. 2. All the measured lines and their assignments are listed in Table 1. Their frequencies were fit using Pickett's program SPFIT [21], to the rotational constant B 0' the centrifugal distortion constant D 0, the nuclear quadrupole and nuclear spin rotation coupling constants, eQq and C ± , for both the AI and N nuclei, and the nuclear spin-spin constant, OrAl_N [22]. The lines observed for the overlapped hyperfine components of the J = 1-0 transition were fit as

K.A. Walker, M. C.L. Gerry/Chemical Physics Letters 278 (1997) 9-15

11

FI = 7 / 2 - 5 / 2

F,: 5/2-5/2

I F,=3/2-5/2

7

Fig. 1. Composite spectrum of the J = 1-0 rotational transition of AINC. The results of three different microwave experiments were used to produce this composite. Each experiment consisted of 200 averaging cycles. 4 K data points were collected at 50 ns sampling interval and transformed.

b l e n d e d lines u s i n g p r e d i c t e d i n t e n s i t i e s as w e i g h t i n g factors. T h e r e s u l t i n g c o n s t a n t s are listed in T a b l e 2. T a b l e 2 also c o m p a r e s the p r e s e n t results w i t h t h o s e o f R o b i n s o n et a l . [ l l ] . T h e r o t a t i o n a l a n d c e n t r i f u g a l d i s t o r t i o n c o n s t a n t s a g r e e quite well,

45 k H z

F=

within one standard deviation. Our high resolution F T M W study p r o d u c e d a m o r e precise B 0. H o w e v e r the v a l u e o f D o d e r i v e d f r o m the m i l l i m e t e r w a v e study was m o r e p r e c i s e b e c a u s e r o t a t i o n a l t r a n s i t i o n s up to J = 31 w e r e m e a s u r e d , w h e r e d i s t o r t i o n effects

F = 9/2 - 7/2

7/2 - 5/2 7/2-7/2

I ~ , F = 5/25/2 5/2- 7/2 5/2- 3/2 _. I

11970.535 ~

k_

.,_J

_~ I

11971.585 M I ~

Fig. 2. Detail of the ] = 1-0 transition of AINC, showing the F~ = 7 / 2 - 5 / 2 hyperfine transition. This spectrum was obtained with 75 averaging cycles. 4 K data points were measured and the power spectrum is displayed with an 8 K transformation. The microwave excitation frequency was 11970.835 MHz.

12

K.A. Walker, M.C.L. Gerry~ Chemical Physics Letters 278 (1997) 9-15

Table 1 Measured frequencies of J = 1-0 and J = 2-1 transitions of AINC Transition J' 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

F'~ 5/2 5/2 5/2 7/2 7/2 7/2 3/2 3/2 3/2 5/2 7/2 7/2 7/2 3/2 3/2 9/2 9/2 7/2 9/2 7/2 7/2 1/2 5/2 5/2 5/2 3/2 3/2

F'

J"

F I'

F"

1/2 7/2 5/2 7/2 9/2 5/2 3/2 1/2 5/2 7/2 5/2 9/2 7/2 5/2 3/2 9/2 11/2 7/2 7/2 9/2 5/2 3/2 5/2 7/2 3/2 3/2, 1/2 5/2

0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

5/2 5/2 5/2 5/2 5/2 5/2 5/2 5/2 5/2 3/2 7/2 7/2 7/2 3/2 3/2 7/2 7/2 5/2 7/2 5/2 5/2 3/2 5/2 5/2 5/2 5/2 5/2

5/2, 5/2, 5/2, 5/2, 7/2 5/2, 5/2, 3/2 5/2, 5/2 5/2 9/2 7/2 5/2 3/2 7/2 9/2 5/2 5/2 7/2 3/2 5/2 5/2 7/2 3/2 5/2, 7/2

are larger. It is i n t e r e s t i n g that in the p r e s e n t w o r k a v a l u e o f D o p r e c i s e to t h r e e figures c o u l d b e obtained. T h i s too reflects the h i g h p r e c i s i o n o f F T M W spectroscopy. Table 2 Molecular constants calculated for AINC Parameter

Value b

B0 Do HoX10 v eQq(27Al) eQq(14 N) C ± (27A1) C± (14N)

5984.67681 (23) 0.003898(30)

CeAI- N

a

Lit. c 5984.6752(43) 0.0038870(73) 0.243(38)

- 35.6268(16) - 2.1508(19) 3.850(84) 1.56(25)

3/2 7/2 7/2, 3 / 2 7/2 7/2, 3 / 2 3/2 7/2, 3 / 2

3/2

Frequency (MHz)

Obs.- calc. (kHz)

11963.2831 11963.4981 11963.9959 11970.8348 11971.2387 11971.2646 11974.2358 11974.4227 11974.4597 23930.8749 23932.3001 23932.3737 23932.6368 23935.9314 23936.5199 23939.2513 23939.4034 23939.4758 23939.5286 23940.1152 23940.2807 23940.5822 23941.6089 23941.8364 23941.9025 23946.7609 23946.8933

- 0.1 -0.0 0.0 -0.1 -0.5 0.1 0.1 0.7 -0.1 0.4 0.8 - 0.5 - 0.0 -0.3 -0.7 0.1 - 0.3 -0.1 0.7 -0.0 0.1 0.1 0.1 0.2 -0.4 -0.1 - 0.1

T h e B o rotational c o n s t a n t also c o m p a r e s fairly well w i t h the t h e o r e t i c a l v a l u e s o f M a et al.[5] as s h o w n in T a b l e 3. It a p p e a r s that the T Z 2 P + f C I S D level o f c a l c u l a t i o n s p r o v i d e d the b e s t e s t i m a t e o f the r o t a t i o n a l c o n s t a n t , to w i t h i n 17 M H z o f the e x p e r i m e n t a l result. T h e p r e d i c t e d B e l a b e l l e d b y M a et al. as the m o s t reliable, h o w e v e r , p r o v e d to b e a b o u t 100 M H z l o w e r in f r e q u e n c y . T h e a s s u m e d q u a d r u p o l e c o u p l i n g c o n s t a n t s , d e r i v e d f r o m A1F a n d M g N C , w e r e f o u n d to p r o v i d e e x c e l l e n t e s t i m a t e s o f

Table 3 Comparison of experimental and theoretical rotational constants

-- 1.27(43)

a Rotational, centrifugal distortion and nuclear quadrupole coupling constants in units of MHz. Nuclear spin--rotation and spinspin constants in units of kHz. b One standard deviation in parentheses, in units of least significant digit. c Results from millimeter wave study [11].

B (MHz) Bo exp. B~ TZ2P + f CISD Be TZ2P + f CCSD(T)

5984.6768 5968 a 5882 a

a From Ref. [5]. Be was calculated since it was determined that B~ and Bo differ by less than 5 MHz.

K.A. Walker, M.C.L. Gerry/Chemical Physics Letters 278 (1997) 9-15 those of A1NC, thereby showing the similarities between these three molecules.

4. Discussion When considering the experimental results for A1NC, comparisons can be made with two types of compounds: the linear metal isocyanides and the aluminium halides. The linear isocyanide structure is preferred for lithium [23] and the alkaline earth [24] as well as aluminium [5] cyanides. It has been observed that the ionic character of the M - N C bond decreases as the calculated energy difference between the linear isocyanide and cyanide structures of these molecules increases [5]. Since AINC has the greatest calculated energy difference of the linear metal isocyanides, the A1-NC bond should be less ionic than that of the alkaline earth and alkali metal isocyanides. Also the properties of A1NC should be comparable to those of AIF and AICI, since the electronegativity of the CN group is similar to those of F and C1. Information obtained from the nuclear quadrupole coupling and nuclear spin-spin constants can be used to determine whether A1NC follows these predicted bonding trends. The nuclear quadrupole coupling constants can be interpreted in terms of valence p-shell electrons using the Townes-Dailey model [25]. This relates the measured molecular quadrupole coupling constant to the atomic coupling constant of an np-electron,

eQq(mol)=

nz

2

eQqn'°(at°m)'

n z = 2a2(1 - a ~ ) + Na~,

where the first term is from the sp-hybrid bonding orbital and the second from the counterhybridised orbital, which may or may not be involved in another bond. In this expression, a 2 is the fractional weight of the sp hybrid in the A I - N bond molecular orbital, a 2 is the s-character of the hybrid orbital and N is s the number of electrons in the counterhybridised orbital. The a 2 value can be estimated from the ionic character of the molecule, i~. For all species discussed, this has been calculated using electronegativities following the expression from Ref. [26], I xA1 - xNc I

ic

2

(2)

'

(3)

where the electronegativity values, x, are taken from Ref. [28]. The ,w-hybridisation of the A1 bonding orbital in several A I - X compounds will be considered first. The valence shell configuration of AI is 3 s 2 3 p ~. The value of 2 a 2 is (1 - i,:), since A1 is the positive pole of the A1-X bond [28]. N will be taken to be 2, because the counterhybridised orbital will have an unshared pair, and the 3px and 3py orbitals are unfilled, so n x = ny = 0. By combining Eqs. (1) and (2) and substituting, the following expression results:

eOq(27 A1) =[(1-i~)(1-a~)+Za~]eQq3L0(Z7Al).

(4)

A further correction must be added to account for decreased nuclear screening of the positive A1 nucleus,

eQq(27A1) = [(1 - i c ) ( 1 - a 2) + 2a~]

(1)

where nx, ny and n~ are the number of electrons in the npx, npy and npz orbitals, respectively. By estimating values for n~, n v and n z, this model can be used to probe orbital hybridization [26-28]. In the case of A1NC, there is a significant contribution to the coupling constants from orbital hybridization on both A1 and N nuclei. We will consider an spz-hybrid orbital on either A1 or N; d orbital contributions will be neglected. In this discussion of linear molecules, the z-axis will be taken to be the molecular axis. The number of pz-electrons in the two hybrid orbitals can be calculated from

13

x ( 1 + ic~)eQq31o(ZTA1),

(5)

where e = 0.35 for AI [29]. Eq. (5) was used to estimate the s-character of the sp-hybridised bonding orbital of A1 for A1NC, AIF and AIC1 and the results are listed in Table 4. These three species are quite Table 4 s-character of Al bonding orbital calculated from Al nuclear quadrupole coupling constants Species eQq(Z7A1)(MHz) a~ AIF AINC A1CI

- 37.49 ~ - 35.627 -30.408 b

0.37 0.26 0.22 ~

From Ref. [19]. u From Ref. [33]. c Differs from value in Ref. [28], because of the method used to calculate i c.

14

K.A. Walker. M. C.L. Gerry / Chemical Physics Letters 278 (1997) 9-15

Table 5 C o m p a r i s o n o f s - c h a r a c t e r o f N b o n d i n g orbital c a l c u l a t e d for several linear i s o c y a n i d e species

eQq( 14N ) ( M H z )

AINC

MgNC a

CaNC b

LiNC c

- 2.151

- 2.323

- 2.697

- 2.941

0.61

0.64

0.62

a~

0.59

a F r o m Ref. [20]. b F r o m Ref. [34]. c F r o m Ref. [23].

ionic compounds (completely so in the case of A1F) and without contributions from sp-hybridization the AI quadrupole coupling constants would be be negligible. The s-character calculated for AINC is intermediate between that of A1F and AIC1 and the values of both eQq(27A1)anda~2 decrease from F to NC to C1. In the same way, an expression can be derived for the sp-hybridization of the N atom in a linear metal isocyanide compound, M - N C . Since the N is on the negative pole of the bond, 2 a 2 is equal to (1 + ic). Care must be taken in determining the p-electron distribution because N is part of the NC ligand and the counterhybridised orbital forms the N - C o- bond. The bonding in the N = C group is assumed to be completely covalent, so each of the pure 2px and 2p~. orbitals and the sp counterhybridised orbital have one half of a bond pair (n X= n~. = N = 1). Once again using Eqs. (1) and (2) and substituting the above values, the following equation results

eQq('4N) =

=

[(l+ic)(1-a~)+

2a~-~--l+l]

× eQq21o(14N) ic(1 - a~)eQq21o('4N).

(6)

In this case, the atomic quadrupole coupling constant must be corrected for the screening effect of the negative charge on the NC group, with E = 0.3 [29],

of the M - N C bond, which increases from A1NC to LiNC. As the M - N C bond becomes more ionic, more electron density is transferred from the metal atom to the N, thereby increasing the magnitude of the quadrupole coupling constant, as predicted by Eq. (6). From the nuclear spin-spin coupling constants, an estimate of the A1-N bond distance can be made. By assuming that the coupling arises from only a direct dipole-dipole interaction, CeA~_N can be expressed as [22]: - 3/-~2 gAl gN OLAI- N

/'31 - N

(8)

This equation uses the nuclear g-factors for A1 and N and the nuclear magneton, /z N. From Eq. (8), a value of rAl_N = 1.75(20) A was calculated. The large uncertainty comes from the relatively large uncertainty in C~AI N. This result compares well, within the uncertainty, with the values calculated by Ma et al.[5]. Also this A1-N bond distance is intermediate between the corresponding distances in A1F [ r e = 1.65436 A [30,31]] and A1C1 [r e = 2.13011 [32]], parallelling the bond properties obtained from the A1 quadrupole coupling constants and again showing the similarities between these three A1-X species,

eQq21o(14N) eQq(14N)=ic(1-a~)

(1 + i c e )

(7)

The value of a~ was calculated for the bonding orbital of N for A1NC and several other linear isocyanides using Eq. (7) and the results are listed in Table 5. Since s-character values for the various linear isocyanides are approximately equal (within ~ 8%), the bonding orbital in each of the isocyanide groups must be similar. The differences in the observed nuclear quadrupole coupling constants appear to arise solely from differences in the ionic character

5. C o n c l u s i o n

Laser ablation has been shown to be an effective route to produce metal isocyanides for investigation by FFMW spectroscopy. This first measurement of the hyperfine structure of A1NC has produced several new parameters, namely the nuclear quadrupole and nuclear spin-rotation constants for both the N and A1 nuclei, and the nuclear spin-spin constant. The sp-hybridisation in the bonding orbitals of AINC

K.A. Walker, M.C.L. Gerr},/ Chemical Physics Letters 2 78 (1997) 9-15

have been investigated through the nuclear quadrupole coupling constants. The orbital hybridisation calculated for N in the linear metal isocyanides is quite similar. The A1NC bond properties have been found to be intermediate between those of AlF and AICI.

Acknowledgements This work has been supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and by the Petroleum Research Fund administered by the American Chemical Society. We acknowledge helpful discussions with J.S. Robinson and L.M. Ziurys during preparation of this paper.

References [1] B.E. Turner, Astrophys. Space Sci. 224 (1995) 297. [2] B.E. Turner, T.C. Steimle, L. Meerts, Astrophys. J. 426 (1994) L97. [3] M. Gu~lin, J. Cernicharo, C. Kahane, J. Gomez-Gonzalez, Astron. Astrophys. 157 (1986) Ll7. [4] K. Kawaguchi, E. Kagi, T. Hirano, S. Takano, S. Saito, Astrophys. J. 406 (1993) L39. [5] B. Ma, Y. Yamaguchi, H.F. Schaefer IIl, Mol. Phys. 86 (1995) 1331. [6] J. Cernicharo, M. Gu~lin, Astron. Astrophys. 183 (1987) LI0. [7] L.M. Ziurys, A.J. Apponi, T.G. Phillips, Astrophys. J. 433 (1994) 729. [8] K.A. Gingerich, Naturwiss. 54 (1967) 646. [9] C. Thomson, Int. J. Quantum Chem. Syrup. 10 (1976) 85. [10] S. Petrie, J. Phys. Chem. 100 (1996) 11581.

15

[l l] J.S. Robinson, A.J. Apponi, L.M. Ziurys, Chem. Phys. Lett. (1997), submitted. [12] M. Fukushima, 52nd Ohio State University International Symposium on Molecular Spectroscopy (1997), Paper MH09. [13] K.A. Walker, M.C.L. Gerry, J. Mol. Spectrosc. 182 (1997) 178. [14] T.J. Balle, W.H. Flygare, Rev. Scient. Instrum. 52 (1981) 33. [15] Y. Xu, W. J~iger, M.C.L. Gerry, J. Mol. Spectrosc. 151 (1992) 206. [16] Y. Xu, W. Jiiger, M.C.L. Gerry, I. Merke, J. Mol. Spectrosc. 160 (1993) 258. [17] J. Haekel, H. Milder, Z. Natur[brsch. Teil A 43 (1988) 203. [18] K. ishii, T. Hirano, U. Nagashima, B. Weis, K. Yamashita, Astrophys. J. 410 (1993) L43. [19] R. Honerjiiger, R. Tischer, Z. Naturlbrsch. Teil A 29 (1974) 342. [20] K.A. Walker, M.C.L. Gerry, 52nd Ohio State University International Symposium on Molecular Spectroscopy (1997), Paper MH06. [21] H.M. Pickett, J. Mol. Spectrosc. 148 (1991) 371. [22] C. Styger, M.C.L. Gerry, J. Mol. Spectrosc. 158 (1993) 328. [23] J.J. van Vaals, W.L. Meerts, A. Dymanus, Chem. Phys. 82 (1983) 385. [24] C.W. Bauschlicher Jr., S.R. Langhoff, H. Partridge, Chem. Phys. Lett. 115 (1985) 124. [25] C.H. Townes, B.P. Dailey, J. Chem. Phys. 17 (1949) 782. [26] W. Gordy, Faraday Discuss. Chem. Soc. 19 (1955) 14. [27] D.R. Lide Jr., J. Chem. Phys. 42 (1965) 1013. [28] W. Gordy, R.L. Cook, Microwave Molcular Spectra, no. XVIII in Techniques of Chemistry (Wiley, New York, 1984) 3rd ed. [29] C.H. Townes, A.L. Schawlow, Microwave Spectroscopy (Dover, New York, 1975). [3(I] F. Wyse, W. Gordy, E.F. Pearson, J. Chem. Phys. 52 (1970) 3887. [31] J. Hoeft, F.J. Lovas, E. Tiemann, T. Ttirring, Z. Naturforsch. Tell A 25 (1970) 1029. [32] F.C. Wyse, W. Gordy, J. Chem. Phys. 56 (1972) 2130. [33] K.D. Hensel, C. Styger, W. Jiiger, A.J. Merer, M.C.L. Gerry, J. Chem. Phys. 99 (1993) 3320. [34] C.T. Scurlock, T.C. Steimle, R.D. Suenram, F.J. Lovas, J. Chem. Phys. 100 (1994) 3497.