Microwave spectrum and structure of the polar N2O dimer

Journal of Molecular Spectroscopy 251 (2008) 153–158

Contents lists available at ScienceDirect

Journal of Molecular Spectroscopy journal homepage: www.elsevier.com/locate/jms

Microwave spectrum and structure of the polar N2O dimer Nicholas R. Walker a,*, Andrea J. Minei b, Stewart E. Novick b, Anthony C. Legon a a b

School of Chemistry, University of Bristol, Bristol BS8 1TH, UK Department of Chemistry, Wesleyan University, Middletown, CT 06459, USA

a r t i c l e

i n f o

Article history: Received 22 January 2008 In revised form 15 February 2008 Available online 29 February 2008 Keywords: van der Waals complex N2O dimer FTMW spectroscopy

a b s t r a c t Cavity Fourier-transform microwave spectroscopy has been used to characterise a gas phase, polar dimer of N2O. The polar (N2O)2 unit is generated by co-expansion of a gas sample containing a small percentage of N2O in helium backing gas. Transitions in the pure rotational spectra of (15N2O)2, (14N15NO)(15N2O), (14N2O)(15N2O) and (14N2O)2 are reported. The measured transitions of (15N2O)2 and (14N15NO)(15N2O) are assigned and fitted to Hamiltonians allowing rotational, centrifugal distortion and 14N nuclear quadrupole coupling constants to be determined. Hyperfine structure is assigned for a single J0K 0 K 0 ! J00K 00 K 00 1 þ1 1 þ1 transition of both isotopomers of (14N2O)(15N2O). Nuclear quadrupole coupling constants, vbb, are reported for all four 14N nuclei. The measured vbb are in excellent agreement with those structures predicted from the measured rotational constants. The geometry of the molecule is slipped-parallel. The sepð1Þ aration between the central nitrogen nuclei of the monomers in the rm structure is 3.570(12) Å with the two N2O monomers, respectively, oriented 54.69(68)° and 49.85(64)° to the a-inertial axis. Simulation of hyperfine structure in the spectrum of the (14N2O)2 isotopomer yields good qualitative agreement with experiment. Ó 2008 Elsevier Inc. All rights reserved.

1. Introduction Early spectroscopic experiments on gas phase N2O clusters confirmed that predissociation occurs following vibrational excitation of (N2O)2 in the ground electronic state [1–3]. Subsequent theoretical studies [4,5] sought to interpret these results without experimental confirmation of the molecular structure. Rotationally-resolved infrared spectroscopy later revealed a non-polar, slipped-parallel structure for a structural isomer of (N2O)2 with the oxygen atoms occupying the ‘‘interior” positions [6–11]. Miller and co-workers have suggested that the nature of the N2O dimer is significant to our understanding of the bulk material [6]. The solid possesses residual entropy explained by a lack of discrimination between ‘‘head-to-head” and ‘‘head-to-tail” pairing of monomers. A description of the structure and binding between N2O monomers is also important from the perspective of our understanding of van der Waals complexes more generally. Many complexes that contain an N2O molecule bound to other small molecules have been studied [12–15]. Two species similar to the N2O dimer, (OCS)2 and (OCO)2, have also been the subject of extensive investigation. Lobue et al. [16] suggested either a T-shaped geometry or a polar, slipped-parallel structure for (OCS)2 following a deflection study in the inhomogeneous electric fields in a molecular beam electric resonance * Corresponding author. Fax: +44 117 925 0612. E-mail address: [email protected] (N.R. Walker). 0022-2852/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jms.2008.02.015

spectrometer. Walsh et al. later observed a non-polar isotopologue [17] of this species by rotationally-resolved infrared spectroscopy and identified the structure of the unit as slipped-parallel. The geometry of (OCO)2 was confirmed as slipped-parallel through experiments performed by the same group [18]. Each of the above studies was performed on complexes initially generated and stabilised through co-expansion of the gaseous monomer in an argon backing gas. Each of these species possesses a molecular structure that does not have a permanent electric dipole moment and hence cannot be probed by microwave spectroscopy. Polar isomers of (N2O)2 and (OCS)2 have only recently been observed by McKellar, Moazzen-Ahmadi and co-workers following co-expansion of the gaseous monomers in helium backing gas [19,20]. The species were identified using rotationally-resolved infrared spectroscopy to reveal polar structures in slipped-parallel geometries. A further study of two isotopologues of (N2O)2 by rotationally-resolved infrared spectroscopy has since allowed a more complete determination of the molecular structure [21]. The determined parameters are in agreement with the results of a recent ab initio study [22]. The structure of the polar OCS dimer was recently determined with improved accuracy through microwave spectroscopy [23]. The work reported here presents the first study of a polar dimer of N2O by the same method. Note that two isotopomers are possible for any given isotopologue of (N2O)2 where the N2O monomers are inequivalent in respect of their isotopic composition. Transitions in the spectra of six different isotopomers are reported with unambiguous assignments provided for the hyperfine structure in

154

N.R. Walker et al. / Journal of Molecular Spectroscopy 251 (2008) 153–158

five of the spectra. All rotational constants have been precisely evaluated for three isotopomers allowing a determination of r0 ð1Þ and r m structures for this molecule. All four vbb(14N) nuclear quadrupole coupling constants are accurately measured allowing a very sensitive examination of the angle between the monomers and the inertial axes. 2. Experimental The Balle–Flygare, Fourier-transform microwave (FTMW) spectrometers [24] used for this work have been described previously [23,25]. The initial experiments on (14N2O)2 were carried out on the FTMW spectrometer at Wesleyan University [26]. The experiments on the five 15N-containing isotopomers were carried out on the FTMW spectrometer at the University of Bristol. A gas sample containing 0.5% N2O in helium undergoes supersonic expansion into the Fabry-Perot cavity through an orifice in the centre of the stationary mirror. All isotopologues studied during this work contained 16O and isotopic substitution exclusively involved the replacement of one or more of the nitrogen atoms within the molecule. For this reason, references to different isotopomers within this work will specify the nitrogen nuclei explicitly unless the general case of the dimer is being cited. Two isotopomers of (15N14NO)(15N15NO) were generated and studied using a mixed sample containing 0.25% of 15N14NO (Amersham International, 96% pure) and 0.25% of 15N2O (Amersham International, 99% pure) in helium gas. (15N2O)2 and (14N2O)2 were each generated from samples containing 0.5% of the pure monomer in helium. Two isotopomers of (14N2O)(15N2O) were generated from a mixed sample containing 0.25% 15N2O and 0.25% 14N2O in helium. Owing to the limited quantity of available sample, measurements were focussed on selected transitions that allowed key parameters to be determined. The use of a gas sample containing >95% helium has previously been found to be essential for the observation of polar dimers of OCS and N2O [19–21,23]. The parallel propagation of the gas and microwave pulses enhances the resolution and sensitivity of the instrument but also increases the well-known Doppler splitting of each hyperfine component so that two peaks are observed for all spectral transitions. The line frequency is determined by finding the average of the frequencies of the two Doppler components. Molecules are excited to higher rotational states by the microwave pulse if the transition from the lower state is resonant or nearly resonant with the radiation and subsequently undergo relaxation by spontaneous coherent emission at the transition frequency. The free induction decay is detected and Fourier-transformed to yield a plot of emission intensity against frequency. The experimental cycle described above is repeated and averaged as necessary to obtain a satisfactory signal-to-noise ratio for measured lines. All frequency measurements are referenced to an external source that is accurate to better than 1 part in 1011. Individual hyperfine components have a full width at half height of 5 kHz and their frequencies are consequently measured with an estimated accuracy of ±0.5 kHz.

tions in the spectrum of (15N2O)2. A transition identified within a few MHz of the initial prediction is shown in Fig. 1. The spectrum of (15N2O)2 is simpler than those of isotopomers containing a greater number of 14N nuclei. Transitions of both isotopomers of (15N14NO)(15N2O) were subsequently identified and assigned before transitions in the spectra of two isotopomers of (14N2O)(15N2O) were measured and assigned. All spectroscopic parameters were determined by fitting assigned transitions to Watson’s ‘A’ reduced Hamiltonian using the program provided by Pickett [27]. Transitions in the spectra of both isotopomers of (15N14NO)(15N15NO) were identified after fewer than 50 nozzle pulses. a-type R branch transitions show no resolved hyperfine structure so it is assumed that unperturbed line centres are measured for these transitions. Fig. 2a illustrates the hyperfine structure observed in the 111–000 transition of one isotopomer of (15N14NO)(15N15NO). Fig. 2b provides a comparison with hyperfine structure in the same transition for an isotopomer which contains an additional substitution of 14N for a 15N nucleus in monomer B. A recent study of (N2O)2 by infrared spectroscopy [21] allowed DK to be precisely determined and its value is fixed in all fits presented here. As shown in Table 1, DJ, DJK, dJ and dK centrifugal distortion constants are accurately determined by fits to transitions in the spectrum of (15N2O)2. The small number of J 0K 0 K 0 ! J 00K 00 K 00 tran1 þ1 1 þ1 sitions measured for isotopomers of (15N14NO)(15N15NO) and 14 15 ( N2O)( N2O) precludes the determination of all quartic centrifugal distortion constants for these species. DJ and DJK are determined in fits to the data acquired from both isotopomers of (15N14NO)(15N15NO) but held fixed to the values obtained for (15N2O)2 in fits to the data from (14N2O)(15N2O). dJ and dK are held fixed to the values obtained for (15N2O)2 in fits to data from all isotopomers of (15N14NO)(15N15NO) and (14N2O)(15N2O). As shown in Table 1, values of A, B and C are precisely determined for (15N2O)2 and both isotopomers of (15N14NO)(15N2O). Only the 111–000, transition was observed for each of the isotopomers of (14N2O)(15N2O). A revised prediction of the molecular structure was therefore used to estimate C for both isotopomers of this isotopologue allowing A to be calculated from the experimental data in each case. The hyperfine structure is completely resolved for each and has been assigned to give unperturbed line centre frequencies at A + C  4DJ and all four vbb(14N) nuclear quadrupole coupling constants. These data are presented in Table 2. Both 14N atoms are known to be located on the same N2O monomer in each isotopomer of (14N2O)(15N2O), usefully allowing the nuclear quadrupole coupling constants of N(A,2) and N(B,2) to be

3. Results and discussion 3.1. Spectroscopic analysis Rotational constants provided by Dehghany et al. [19] were used to predict the rotational spectrum of the isotopologue containing the most naturally-abundant monomer (14N2O)2. After a short search, transitions in the spectrum of this molecule were identified. An approximation of the molecular structure was found to yield rotational constants in good agreement with the experimentally-determined quantities and then used to predict transi-

Fig. 1. The 111–000 transition of (15N2O)2 after 450 gas pulses.

155

N.R. Walker et al. / Journal of Molecular Spectroscopy 251 (2008) 153–158

Fig. 2. (a) Hyperfine structure in the 111–000 transition of (15N14NO)(15N2O) where the 14N nucleus is located in the N(B,2) position. The coupling scheme used is J + IN = F and transitions are labelled F0  F00 . The transition is displayed after 10 500 gas pulses. (b) Hyperfine structure in the 111–000 transition of one isotopomer of (14N2O)(15N2O). The coupling scheme used is J + IN(B,1) = FN(B,1) and FN(B,1) + FN(B,2) = F and hyperfine components are labelled F 0NðB;1Þ  F 00NðB;1Þ , F0  F00 . The transition is displayed after 5 000 gas pulses. The peak labelled with an asterisk is the low frequency component of a Doppler pair centred at 10529.6803 MHz. This feature is assigned to the 2–0, 1–0 and 2–2, 1–2 transitions which are too close in frequency to be resolved (see supplementary data).

Table 1 Spectroscopic constants for the all-

14

N and all-

15

N isotopomers of (N2O2)

Constant (MHz)

(14N2O)2 Ref. [19]

(14N2O)2 Ref. [21]

(14N2O)2 [This work]

(15N2O)2 [This work]

(15N2O)2 Ref. [21]

A B C DJ  103 DJK  103 DK dJ  103 dK  103

9269.70(54)a 1621.55(12) 1376.20(9) — — — — —

9269.97(63) 1622.69(18) 1376.65(15) 9.1(13) 43(13) 0.35(7) 1.6(7) —

9268.723(55) 1622.5964(87) 1376.6244(77) 9.1056b 82.021b 0.38c 1.78220b 19.00b

8937.38264(60) 1553.0920(19) 1319.2053(18) 9.1056(93) 82.021(51) 0.38c 1.7822(11) 19.00(93)

8936.54(48) 1553.01(12) 1319.18(9) 8.7(7) 87(9) 0.38(5) 1.8(4) —

a b c

Number in parentheses is one r in units of the last significant figure. Fixed at value determined for (15N2O)2. Fixed at value determined for (15N2O)2 in Ref. [21].

Table 2 Spectroscopic constants for the mixed isotopomers of (N2O)2 Constant (MHz)

A B C DJ  103 DJK  103 DK  103 dJ  103 dK  103 vbb(N(A,1))  103 vbb(N(A,2))  103 vbb(N(B,1))  103 vbb(N(B,2))  103 m(111–000) a b c d

(15N14NO)(15N2O)

(14N2O)(15N2O)

N(A,2)a

N(B,2)a

N(A,1), N(A,2)a

N(B,1), N(B,2)a

8944.17229(78) 1569.02049(78) 1330.83556(46) 9.252(10) 82.52(34) 0.38d 1.78220c 19.00c — 85.3(13) — — 10274.9708(13)

8949.9350(29) 1567.8965(29) 1330.1490(17) 9.2444(40) 83.4(13) 0.38d 1.78220c 19.00c — — — 136.0(50) 10280.0470(46)

9011.24200(56) 1599.7550b 1357.2870b 9.1056c 82.021c 0.38d 1.78220c 19.00c 277.8(24) 86.8(84) — — 10368.4926(56)

9188.33981(71) 1572.0940b 1340.9791b 9.1056c 82.021c 0.38d 1.78220c 19.00c — — 368.4(34) 129.4(58) 10529.2825(71)

Indicates which of the nuclei are 14N using the convention introduced in Fig. 3. ð1Þ Fixed at values predicted from rm structure. Fixed at value determined for (15N2O)2. Fixed at value determined for (15N2O)2 in Ref. [21].

156

N.R. Walker et al. / Journal of Molecular Spectroscopy 251 (2008) 153–158

distinguished from each other. Thus, all four vbb(14N) nuclear quadrupole coupling constants are unambiguously determined. Assignment of hyperfine structure in the spectrum of (14N2O)2 would present no further information and could not be achieved with a comparable degree of accuracy. For this reason, the hyperfine parameters presented herein were exclusively determined using the data acquired from isotopomers that contain either one or two 14N nuclei.

Table 3 Structural parameters for (N2O)2

r0 ð1Þ rm r0 Ref. [21] Ab initio Ref. [22] a b

r(N(A,2)  N(B,2))a

h Aa

hBa

3.613(5) 3.570(12)b 3.61(14) 3.442

61.1(29) 62.76(68) 61.0(25) 60.6

58.4(28) 57.92(64) 58.5(25) 60.5

r(N(A,2)  N(B,2)) in Å, hA and hB in degrees. ca = 0.13(11), cb = 0.41(12), cc = 0.39(10).

3.2. Molecular structure determination The fits yield precisely determined A, B and C rotational constants for (15N2O)2 and both isotopomers of (15N14NO)(15N2O). The rotational constants of (14N2O)2 are not determined as precisely owing to the extensive hyperfine structure in the spectrum of this isotopologue. However, the observation of 16 transitions allows these quantities to be approximated. Thus, 12 experimentally-determined rotational constants are available to fit the structure of the polar N2O dimer. The inertial defect of polar (14N2O)2 can be calculated to be 1.13 u Å2 using the experimentally-determined rotational constants indicating that the molecule is almost certainly planar. The quantity for the non-polar dimer is 0.65 u Å2. The r(N(1)  N(2)) and r(N(2)  O) distances within the dimer can be assumed to be unchanged from those in the N2O monomer. As shown in Fig. 3, the remaining structural parameters can therefore be defined with respect to three quantities. These include the distance between the central nitrogen atoms, r(N(A,2)  N(B,2)) and the two angles, hA and hB, between the monomer axes and the line connecting the two interior nitrogen atoms. For the purposes of this work, it is necessary to establish a convention that allows different isotopomers of the same isotopologue to be distinguished explicitly within the text. The distinct positions occupied by N2O monomers are therefore labelled ‘‘A” and ‘‘B” while the distinct positions of nitrogen nuclei within each monomer are labelled ‘‘1” and ‘‘2” in the figure. Individual N nuclei are thus distinguished explicitly here using the logical convention that N(A,1) indicates a 14N nucleus in monomer A at position 1. A least-squares fit of these three structural parameters to the 12 rotational constants was performed using Kisiel’s STRFIT [28] program. The results are shown in Table 3. The r0 structure of the dimer [29] includes r(N(1)  N(2)) and r(N(2)  O) fixed equal to the r0 values of the 14N14N16O monomer

which are 1.1278 and 1.1923 Å, respectively. The determined quantities are r(N(A,2)  N(B,2)) = 3.613(5) Å, hA = 61.1(29)° and hB = 58.4(28)°. All of the evaluated parameters are in excellent agreement with the most recent study by rotationally-resolved ð1Þ infrared spectroscopy. The r m structure of the dimer assumes ð2Þ r(N(1)  N(2)) and r(N(2)  O) bond lengths appropriate to the rm structure of the monomer [30] which are 1.1268 and 1.1855 Å, ð1Þ respectively. The value of r(N(A,2)  N(B,2)) yielded by the r m structure is slightly shorter than the equivalent quantity in the r0 structure and is not as precisely determined. Conversely, both angles are ð1Þ more precisely determined in the rm structure. Both the values determined for r(N(A,2)  N(B,2)) by the above fitting methods are greater than a recent ab initio result [22] by 0.15 Å but very good agreement is not necessarily expected here. The ab initio calculation predicts an re geometry while the microwave data permit the calculation of a structure for the vibrational ground state only. 3.3. Nuclear quadrupole coupling constants Simple expressions connect the experimentally-determined values of vbb for (N2O)2 with the molecular structure and the nuclear quadrupole coupling constants of the individual N2O monomer units. It is thus possible to test whether structures established through least squares fitting of the rotational constants are consistent with information available from the nuclear quadrupole coupling constants. The values determined for hA and hB allow calculation of the angles between the monomers and the inertial axes of the molecule, aA and aB, as shown in Table 4. If changes in electric field gradients at 14N nuclei relative to the monomer and zero-point averaging effects in the dimer are both ignored, the nuclear quadrupole coupling constants of the dimer, vaa, vbb and vcc are related to those of the monomer [31], v0, as follows: 1 ð3 cos2 a  1Þv0 2 1 2 vbb ¼ ð3 sin a  1Þv0 2 1 vcc ¼  v0 2

vaa ¼

ð1Þ ð2Þ ð3Þ

Table 4 Nuclear quadrupole coupling constants calculated from structural fits and from the spectra

Fig. 3. The structure of (N2O)2 determined from the observed rotational constants. The distinct positions occupied by N2O monomers are labelled ‘‘A” and ‘‘B” while the distinct positions of nitrogen nuclei within each monomer are labelled ‘‘1” and ‘‘2”. Individual N nuclei are thus distinguished explicitly within the text using the logical convention that N(A,1) indicates a 14N nucleus in monomer A at position 1. Monomers A and B are not equivalent leading to four distinct 14N nuclear quadrupole coupling constants in the (14N2O)2 isotopologue. The angles between the monomer axes and the inertial axes of the molecule are aA = 54.69(68)° and aB = 49.85(64)°, ð1Þ respectively, in the r m structure. The separation between the N(2) nuclei of the two monomers is determined to be 3.570(12) Å by the same method.

a

r0 ð1Þ rm

vaa (kHz)

Calc. Calc. Calc. Calc. Obs. Calc.

vbb (kHz)

vcc (kHz)

(r0) ð1Þ ðrm Þ (r0) ð1Þ ðrm Þ

aA/°

aB/°

53.2(29) 54.69(68)

50.5(28) 49.85(64)

N(A,1)

N(A,2)

N(B,1)

N(B,2)

29.6 0.87 357.3 386.0 368.4(34) 386.9

9.7 0.30 123.6 133.5 136.0(50)a 133.8

82.7 95.7 304.2 291.2 277.8(24) 386.9

28.6 33.1 105.2 100.7 85.3(13)a 133.8

a The quantities given are those determined in fits of the spectra of (15N14NO)(15N2O) isotopomers. Comparison with the equivalent quantity determined by fitting the spectra of (14N2O)(15N2O) in Table 2 reveals a similar level of agreement with calculation.

N.R. Walker et al. / Journal of Molecular Spectroscopy 251 (2008) 153–158

where the value of a can be chosen to correspond with the value appropriate to either monomer A or monomer B. v0(14N(1)) for the N2O monomer is 773.76(27) kHz and v0(14N(2)) is 267.58(38) kHz [31]. Table 4 shows the results of a comparison between the experimentally-observed nuclear quadrupole coupling constants and those calculated from the structural fits. As described previously, hyperfine structure was resolved only for the 111–000, b-type transitions and neither vaa nor vcc was determined experimentally. The experimentally-observed vbb generally deviate from the ð1Þ quantities calculated from the rm structure by less than 10%. The r0 structure is less consistent with the experimental results but remains accurate within about 20% of the magnitude of the parameter. In agreement with the experimental observation that hyperfine splittings are not resolved in a-type transitions of the two isotopomers of (15N14NO)(15N15NO), vaa(N(A/B,2)) is predicted to be less than 35 kHz for both monomers. It should be noted that the magnitudes of the nuclear quadrupole coupling constants are predicted to be highly sensitive to the monomer angles at the molecular geometry determined from the rotational constants. The level of agreement between the experimental quantities and those predicted from the results of the structural fits is very satisfying. 3.4. Simulation of transitions for (14N2O)2 The microwave experiments on (N2O)2 started, quite naturally, with the major isotopomer, (14N2O)2. Owing to the four, spin 1, 14N nuclei, these transitions consist of hundreds of unresolved hyperfine components within 1 MHz windows. Even with the eventual assignments of the various hyperfine components of the transitions of the partially and fully 15N-substituted isotopomers of the nitrous oxide dimer, the all nitrogen-14 isotopomer cannot be even partially assigned. The best we can do is simulate the various transitions and compare these with the experimental measurements. ð1Þ Given the rm structure shown in Fig. 3, the quadrupole tensors of the four 14N [31], can be rotated into the principal axis system of the complex. These tensors have vaa, vbb, vcc and vab, non-zero components. Using Pickett’s SPCAT program [27], these quadrupole components can easily be incorporated into a spectral prediction. The 312–303 transition of the (14N2O)2, for example, is predicted to consist of 84 transitions of intensity within a factor of 10 of

157

the strongest line, 782 transitions within a factor of 100 of the strongest line, and 1963 transitions within a factor of 1000 of the strongest line, all within a frequency range of 500 kHz. Each transition (all 2980 transitions predicted by SPCAT), with its appropriate intensity, was assigned a Lorentzian line shape with a FWHM of 5 kHz. Each of these line shapes was added together, assuming no interference, to arrive at the simulated spectrum shown inverted in Fig. 4. Shown in the positive direction in the figure is the experimental spectrum. The agreement is qualitatively pleasing (as are those of the other transitions) but it is not an exact quantitative match. If the structure of the dimer is significantly changed from our experimentally derived structure (say by 10° for a) the simulation no longer mimics the experimental spectra, even qualitatively. The reason for the lack of quantitative agreement between the simulated and experimental spectra is that free induction decays (FIDs) of transitions that overlap in frequency can interfere with one another in rather complex ways [32]. Thus the Fourier-transforms of overlapping transitions do not simply add. For simple overlapping spectra of few lines, the FIDs can be fit directly each with variable phase, frequency, and intensity [33]. This is obviously impossible in the current case. 4. Conclusion Transitions in the microwave spectra of six isotopomers of a polar structural isomer of (N2O)2 have been detected and transitions in five of the spectra have been assigned. The molecular structure of the polar dimer of (N2O)2 has been precisely determined through fits to rotational constants determined for four isotopomers. All four 14N nuclear quadrupole coupling constants have been precisely determined through assignment of hyperfine structure in the spectra of four isotopomers allowing a sensitive test of the molecular geometry. It has thus been confirmed that the angles between the axes of monomers A and B and the inertial a-axis are ð1Þ most precisely determined by the r m structure where they are 54.69(68)° and 49.85(64)°, respectively. Acknowledgments The authors thank Amersham International plc. and Prof. C.L. Willis for providing the isotopic samples necessary for these experiments. N.R.W. thanks the Royal Society for the award of a University Research Fellowship. S.E.N. wishes to thank the Petroleum Research Fund of the American Chemical Society for support. Appendix A. Supplementary data Supplementary data for this article are available on ScienceDirect (www.sciencedirect.com) and as part of the Ohio State University Molecular Spectroscopy Archives (http://library.osu.edu/sites/ msa/jmsa_hp.htm). Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jms. 2008.02.015. References [1] [2] [3] [4] [5] [6] [7]

Fig. 4. The experimental and simulated spectrum for the 312–303 transition of (14N2O)2. The spectrum consists of hundreds of unresolved hyperfine components due to the four, spin 1, 14N nuclei. The simulated spectrum assumes Doppler splittings appropriate to the velocity of the helium beam. See text.

T.E. Gough, R.E. Miller, G. Scoles, J. Chem. Phys. 69 (1978) 1588–1590. R.E. Miller, R.O. Watts, A. Ding, Chem. Phys. 83 (1984) 155–169. L.S. Bernstein, C.E. Kolb, J. Chem. Phys. 71 (1979) 2818–2823. J.A. Beswick, J. Jortner, J. Chem. Phys. 74 (1981) 6725–6733. D.A. Morales, G.E. Ewing, Chem. Phys. 53 (1980) 141–147. Z.S. Huang, R.E. Miller, J. Chem. Phys. 89 (1988) 5408–5416. B.J. Howard, 42nd Symposium on Molecular Spectroscopy, The Ohio State University, Columbus, OH, 1987. [8] H.-B. Qian, W.A. Herrebout, B.J. Howard, Mol. Phys. 91 (1997) 689–696. [9] Y. Ohshima, Y. Matsumoto, M. Takami, K. Kuchitsu, Chem. Phys. Lett. 152 (1988) 294–298. [10] M. Gauthier, J. Chem. Phys. 88 (1988) 5439–5449.

158 [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]

N.R. Walker et al. / Journal of Molecular Spectroscopy 251 (2008) 153–158 A. Hecker, I. Scheele, M. Havenith, Phys. Chem. Chem. Phys. 5 (2003) 2333–2336. G. Gimmler, M. Havenith, J. Mol. Spectrosc. 216 (2002) 315–321. M.D. Marshall, H.O. Leung, J. Mol. Spectrosc. 196 (1999) 149–153. J. Tang, A.R.W. McKellar, J. Chem. Phys. 117 (2002) 8308–8313. W.A. Herrebout, H.-B. Qian, H. Yamaguchi, B.J. Howard, J. Mol. Spectrosc. 189 (1998) 235–243. J.M. Lobue, J.K. Rice, S.E. Novick, Chem. Phys. Lett. 112 (1984) 376–380. R.W. Randall, J.M. Wilkie, B.J. Howard, J.S. Muenter, Mol. Phys. 69 (1990) 839. M.A. Walsh, T.H. England, T.R. Dyke, B.J. Howard, Chem. Phys. Lett. 142 (1987) 265–270. M. Dehghany, M. Afshari, Z. Abusara, N. Moazzen-Ahmadi, A.R.W. McKellar, J. Chem. Phys. 126 (2007) 164310. M. Afshari, M. Dehghany, Z. Abusara, N. Moazzen-Ahmadi, A.R.W. McKellar, J. Chem. Phys. 126 (2007) 071102. M. Dehghany, M. Afshari, Z. Abusara, C. van Eck, N. Moazzen-Ahmadi, J. Mol. Spectrosc. 247 (2008) 123–127. H. Valdés, J.A. Sordo, J. Phys. Chem. A 108 (2004) 2062–2071. This paper reports an ab initio structure for polar (N2O)2. The corresponding author has

[23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]

since indicated in a private communication that the geometrical parameters for the N2O  N2O (c) (Cs) planar structure included in Table 6 of this reference are not correct. The following MP2/cc-pVTZ parameters correspond to the Cs reported structure: r(N(A,2)  N(B,2)) = 3.442 Å, hA = 60.6°, hB = 60.5°, A = 8630.40 MHz, B = 1785.86 MHz and C = 1479.68 MHz. A.J. Minei, S.E. Novick, J. Chem. Phys. 126 (2007) 101101. T.J. Balle, W.H. Flygare, Rev. Sci. Instrum. 52 (1981) 33–45. A.C. Legon, in: G. Scoles (Ed.), Atomic and Molecular Beam Methods, vol. 2, Oxford University Press, Oxford, 1992 (Chapter 9). A.J. Minei, S.E. Novick, 62nd Symposium on Molecular Spectroscopy, The Ohio State University, Columbus, OH, 2007. H.M. Pickett, J. Mol. Spectrosc. 148 (1991) 371–377. Z. Kisiel, J. Mol. Spectrosc. 218 (2003) 58–67. C.C. Costain, J. Chem. Phys. 29 (1958) 864–874. J.K.G. Watson, A. Roytburg, W. Ulrich, J. Mol. Spectrosc. 196 (1999) 102–119. J.M.L.J. Reinartz, W.L. Meerts, A. Dymanus, Chem. Phys. 31 (1978) 19–29. I. Merke, H. Dreizler, Z. Naturforsch. 43a (1988) 196–202. J. Haekel, H. Mäder, Z. Naturforsch. 43a (1988) 203–206.