- Email: [email protected]
High-resolution FTIR spectrum of the ν12 band of ethylene-d (C2H3D)
Journal of Molecular Spectroscopy 228 (2004) 105–109 www.elsevier.com/locate/jms
High-resolution FTIR spectrum of the m12 band of ethylene-d (C2H3D) T.L. Tana,*, K.L. Goha, H.H. Teob a
b
Natural Sciences Academic Group, National Institute of Education, Nanyang Technological University, 1, Nanyang Walk, Singapore 637616, Singapore Department of Physics, Faculty of Science, National University of Singapore, Lower Kent Ridge Road, Singapore 119260, Singapore Received 7 May 2004; in revised form 16 July 2004 Available online 20 August 2004
Abstract The infrared absorption spectrum of the m12 fundamental band of ethylene-d (C2H3D) has been recorded with an unapodized resolution of 0.004 cm1 in the wavenumber range of 1340–1460 cm1 using the Fourier transform technique. By assigning and fitting a total of 870 infrared transitions using a WatsonÕs A-reduced Hamiltonian in the Ir representation, three rotational and five quartic centrifugal distortion constants for the upper state (v12 = 1) were determined for the first time. The rms deviation of the fit was 0.00044 cm1 which is close to the experimental precision of the absorption lines. The A-type m12 band centred at 1400.762811 ± 0.000041 cm1was found to be relatively free from local frequency perturbations. The inertial defect D12 was found ˚ 2. to be 0.20928 ± 0.00002 lA 2004 Elsevier Inc. All rights reserved. Keywords: FTIR spectrum of ethylene-d
1. Introduction In the past decades, low-resolution infrared measurements [1–4] have been made on the ethylene-d (C2H3D) molecule and some of its isotopomers for the understanding of their basic vibrational structures. In the studies of Duncan et al. [1,4], most of the vibrational bands of C2H3D were identified and assigned accurately. Furthermore, infrared measurements and analyses [5] of C2H3D in the 2830–3270 cm1 region with a resolution better than 0.030 cm1 were conducted to study the Fermi resonance between m1 and m5, and to derive rotational and centrifugal distortion constants for m5, m2 + m3, and m9. The pure rotational spectra of C2H3D were observed and analysed by microwave spectroscopy [6]. In this
*
Corresponding author. Fax: +65 6896 9414. E-mail address: [email protected] (T.L. Tan).
0022-2852/$ - see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jms.2004.07.008
work [6], accurate rotational and centrifugal distortion constants in the ground state were determined, and the average structure of C2H3D molecule was also calculated. Most of the recent high resolution (at 0.004 cm1 or better) FTIR studies on the ethylene molecules were focussed on C2H4 [7,8], C2D4 [9–11], trans-C2H2D2 [12,13], cis-C2H2D2 [14,15], and 1,1-ethylene-d2 [16]. However, FTIR spectroscopic works on C2H3D at resolution of 0.004 cm1 or better have not been reported so far. The purpose of this work is to measure and analyse the m12 band of ethylene-d (C2H3D) which was measured with a resolution of 0.004 cm1. By fitting 870 assigned transitions, the three rotational and five quartic centrifugal distortion constants of the v12 = 1 state were obtained for the first time. The rms deviation of the fit of transitions of the A-type m12 band was 0.00044 cm1. As the rms deviation of the fit is close to the estimated absolute accuracy (0.0005 cm1) of the measured
106
T.L. Tan et al. / Journal of Molecular Spectroscopy 228 (2004) 105–109
transition, it can be concluded that the m12 band is relatively free of local frequency perturbations.
2. Experimental details The spectra of the m12 band of C2H3D were recorded using a Bomem DA3.002 Fourier transform spectrophotometer [8–11], with a unapodized resolution of 0.004 cm1. A Globar infrared source, and a high sensitivity liquid nitrogen cooled HgCdTe detector, and KBr beam splitter were used. A low-pass infrared band filter (0–2500 cm1) was used to reduce background noise level. The final spectrum was produced by coadding four runs of 60 scans each with the total scanning time of about 23 h. The linewidth (FWHM) in the final spectrum was found to be about 0.0045 cm1. The ethylene-d (C2H3D) gas samples of chemical purity better than 98%, used in these measurements were supplied by Cambridge Isotope Laboratories (Cambridge, MA). For the spectral measurements at the ambient temperature of about 296 K, we used a single pass 20-cm absorption cell with ZnSe windows. The vapor pressure of about 5 Torr in the cell was measured using capacitance pressure gauge. A spectrum of the evacuated cell, consisting of 200 scans at a resolution 0.1 cm1, was also measured and transformed at 0.004 cm1 with zero-filling to obtain a background spectrum. This spectrum when ratioed with the sample spectrum of C2H3D gave a transmittance spectrum with relatively smooth baseline. The spectrum in the 1330–1460 cm1 region showed some strong H2O absorption lines due to H2O impurity in the gas cell. However, the H2O lines did not cause any interference with those of C2H3D since the line positions of H2O are known accurately [17] and the lines are widely spaced. Calibration of the absorption lines of m12 of C2H3D was conducted using 34 unblended and unsaturated H2O lines in the 1330–1460 cm1 region. The selected H2O calibration frequencies were taken from Guelachvili and Rao [17]. A correction factor of 1.00000096 was required to bring the observed wavenumbers into agreement with the calibrated frequencies. From the line fitting the relative precision of the wavenumbers obtained was in the order of 0.0004 cm1. After considering small systematic errors in the experiments, it is reasonable to estimate the absolute accuracy of the measured C2H3D lines to be approximately ± 0.0005 cm1.
symmetry and is one of the 12 fundamental modes of vibrations [1,4]. The m12 mode is ascribed to an in-plane C–H bending. A plot of the spectrum of m12 recorded with a resolution of 0.5 cm1 in the 1330–1470 cm1 region is shown in Fig. 1. It shows the presence of a prominent strong central Q branch at about 1400.76 cm1, which is typical of an A-type band. In the rotational analysis, the band has been confirmed to be A-type. In the high resolution plot, strong absorption lines were easily observed with a regular spacing of about 1.66 cm1 which is about B + C. During the initial assignment of the P and R branches of the band, the strongest of these lines was assigned as J = Kc and Ka = 0 or 1 which were resolved with decreasing magnitude, up to J 0 = 30 and 27 in the P and R branches respectively. These transitions become unresolved doublets for higher J 0 ¼ K 0c and Ka = 0 or 1 values. Fig. 2 shows the asymmetry splitting for J 0 ¼ K 0c ¼ 4, Ka = 0 and 1 transitions in the J 0 = 4 cluster in the P branch. Asymmetry splitting was also observed for Ka = 2 in the J 0 = 4 cluster, while the transitions of Ka = 3, and 4 were not split and were unresolved doublets, as shown in Fig. 2. The intensity of the transitions in the same J 0 cluster decreases for higher Ka values. As the J 0 value increases the asymmetry splitting occurs for even higher Ka values for both P and R branches. Fig. 3 shows the asymmetry splitting for J 0 ¼ K 0c ¼ 5, Ka = 0and 1 transitions in the J 0 = 5 cluster in the R branch. Asymmetry splitting for Ka = 2 transitions is also observed. Since ground state rotational constants for C2H3D were available (6), initial assignment of the transitions could be accurately made, starting with low J values. Those low J transitions were fit to get preliminary constants that were used to predict the higher J transitions. These transitions were then assigned and the process repeated to extend the assignments to even higher quantum numbers. This bootstrap procedure allowed us to assign all the transitions accurately. Accurate rovibrational band constants up to five quartic terms obtained from the fitting of the P and R branch transitions were used to calculate the Q branch transitions with high precision. A total of 256 well-resolved Q branch transitions were confidently assigned and used in the final fit. The strongest lines were all doublets with two possible values of Kc. There were enough fairly unblended lines in the strong Q branch to ensure the accuracy of the present assignments. Fig. 4 shows the m12 Q branch region with well-defined K 0a clusters. For each K 0a cluster, the line intensity is maximum for J 0 ¼ K 0a and K 0c ¼ 0 or 1, and the intensity decreases as J 0 increases.
3. Assignments and frequency analysis 4. Results and discussion The C2H3D molecule is a simple asymmetric top planar molecule of Cs symmetry with asymmetry parameter j of about 0.89. The m12 band is infrared active with A 0
The nonlinear least-squares program which was previously used in the analyses of the m12 bands of 12C2H4
T.L. Tan et al. / Journal of Molecular Spectroscopy 228 (2004) 105–109
Fig. 1. A survey spectrum with resolution of 0.5 cm1 of the m12 band of C2H3D.
Fig. 2. A detailed section of the P branch region of C2H3D.
107
108
T.L. Tan et al. / Journal of Molecular Spectroscopy 228 (2004) 105–109
Fig. 3. A detailed section of the R branch region of C2H3D.
Fig. 4. A survey spectrum with resolution of 0.004 cm1 in the Q branch region of C2H3D.
T.L. Tan et al. / Journal of Molecular Spectroscopy 228 (2004) 105–109 Table 1 Ground state and upper state (v12 = 1) constants (cm1) for C2H3D (A-reduction, Ir representation)
A B C DJ · 106 DJK · 106 DK · 105 dJ · 107 dK · 106 m0 Number of infrared transitions Rms deviation (cm1) ˚ 2) D12 (lA
Ground state [6]a
v12 = 1
4.0058887(23) 0.91632515(87) 0.74377271(47) 1.3010(80) 6.012(66) 7.0599(53) 2.806(10) 8.114(79) —
4.0390015(12)b 0.92022114(63) 0.74255820(54) 1.36209(54) 4.6673(35) 8.05600(94) 3.0941(41) 9.503(24) 1400.762811(41) 870 0.000 44 0.209280(20)
0.059821(2)
a
All five quartic constants were included in the present fit. The uncertainty in the last digits (twice the estimated standard error) is given in parentheses. b
[8] and 12C2D4 [9] was also employed to fit the assigned transitions of m12 of C2H3D. This program for fitting asymmetric rotor spectra uses a Watson Hamiltonian [18] with an Ir representation in an A-reduction. Ground state rovibrational constants of C2H3D inclusive of all five quartic terms were initially taken from Hirota et al. [6]. These constants represent the most accurate values available for C2H3D. They were derived with high precision from 21 observed transitions in the microwave spectra of C2H3D. The accuracies of all five quartic constants were improved during the fit. In our rotational analysis, the ground state constants were fixed to determine the upper state v12 = 1 constants of C2H3D inclusive of all five quartic terms. Inclusion of sextic constants did not improve the accuracy of the fit. A total of 870 infrared transitions was finally assigned and fitted in the determination of the v12 = 1 rovibrational constants as given in Table 1. The m12 band centre is found to be 1400.762811 ± 0.000041 cm1. This value is smaller than that of 1401.5 cm1 measured by Duncan et al. [1], in 1973, in their study of 13C frequency shifts in ethylene. In a recent work by Duncan et al. [4], in 1993, the band centre was observed to be 1400.0 cm1. The fitting program, though written in terms of prolate matrix elements, had no problem in rapidly converging to the correct solution after several iterations. In the least square analysis, the infrared measurements were weighted by the inverse square of the estimated uncertainty. The infrared transitions were given an uncertainty of 0.0004 cm1 which is close to the absolute accuracy of the measured line. The rms deviation of the fit with all weighted lines was 0.00044 cm1. The transitions include values of Ka ranging from 0 to 12. For the P branch, we have J 0 and K 0c ranging from 0 to 30. For the R branch, J 0 and K 0c values range from 1 to 27. The values of J and Kc range from 1 to 22 and 0 to 13, respectively, for the Q branch. The wide ranging
109
values of J 0 , K 0a , and K 0c used in the analysis effectively cover the whole frequency range of 1340–1460 cm1 for the m12 band of C2H3D. ˚ 2 of C2H3D deThe inertial defect D12 of 0.20928(2) lA rived from the present analysis, as given in Table 1 is less ˚ 2; and that of than that of C2H4 [8] which is 0.24201(2) lA 2 ˚ trans-C2H2D2 [13] which is 0.29967(2) lA ; and also that ˚ 2. The upper of cis-C2H2D2 [15] which is 0.24820(2) lA state rovibrational constants A, B, DJ, DK, dJ, and dK of m12 band of C2H3D as given in Table 1 are found to be greater than the corresponding ground state values, while the constants C and DJK are lower than those of ground states. These trends in sign are in good agreement with those in the m12 band of C2H4 [8]. This result indicates similarity in the rotational structures of the m12 bands of both ethylene molecules. In conclusion, the present investigation reports the measurements, assignments, and fitting of 870 infrared transitions of the m12 band of C2H3D. Accurate upper state (v12 = 1) constants inclusive of all five quartic terms were derived for the first time. These spectral data would be another useful contribution to the understanding of the molecular structure of ethylene. References [1] J.L. Duncan, D.C. McKean, P.D. Mallinson, J. Mol. Spectrosc. 45 (1973) 221–246. [2] Y. Verbist-Scieur, C.P. Courtoy, A. Fayt, D. Van Lerberghe, Ann. Soc. Sci. Brux. 90 (1976) 317–336. [3] A. Fayt, Ann. Soc. Sci. Brux. 84 (1970) 69–106. [4] J.L. Duncan, A.M. Ferguson, S.T. Goodlad, Spectrochim. Acta 49A (1993) 149–160. [5] Y. Verbist-Scieur, C.P. Courtoy, A. Fayt, J. Mol. Spectrosc. 85 (1981) 480–492. [6] E. Hirota, Y. Endo, S. Saito, K. Yoshida, I. Yamaguchi, K. Machida, J. Mol. Spectrosc. 89 (1981) 223–231. [7] R. Georges, M. Bach, M. Herman, Mol. Phys. 90 (1997) 381–387. [8] T.L. Tan, S.Y. Lau, P.P. Ong, K.L. Goh, H.H. Teo, J. Mol. Spectrosc. 203 (2000) 310–313. [9] T.L. Tan, K.L. Goh, P.P. Ong, H.H. Teo, Chem. Phys. Lett. 315 (1999) 82–86. [10] K.L. Goh, T.L. Tan, P.P. Ong, H.H. Teo, Mol. Phys. 98 (2000) 583–587. [11] T.L. Tan, K.L. Goh, P.P. Ong, H.H. Teo, J. Mol. Spectrosc. 202 (2000) 249–252. [12] F. Hegelund, J. Mol. Spectrosc. 135 (1989) 45–58. [13] H.H. Teo, P.P. Ong, T.L. Tan, K.L. Goh, J. Mol. Spectrosc. 204 (2000) 145–147. [14] F. Hegelund, F.M. Nicolaisen, J. Mol. Spectrosc. 128 (1988) 321– 333. [15] K.L. Goh, T.L. Tan, P.P. Ong, H.H. Teo, Chem. Phys. Lett. 325 (2000) 584–588. [16] F. Hegelund, J. Mol. Spectrosc. 148 (1991) 415–426. [17] G. Guelachvili, K. Narahari Rao, Handbook of Infrared Standards, Academic Press, Orlando, FL, 1986. [18] J.K.G. Watson, in: J.R. Durig (Ed.), Vibrational Spectra and Structure, A Series of Advances, vol. 6, Elsevier, New York, 1977 Chapter 1.
High-resolution FTIR spectrum of the m12 band of ethylene-d (C2H3D) T.L. Tana,*, K.L. Goha, H.H. Teob a
b
Natural Sciences Academic Group, National Institute of Education, Nanyang Technological University, 1, Nanyang Walk, Singapore 637616, Singapore Department of Physics, Faculty of Science, National University of Singapore, Lower Kent Ridge Road, Singapore 119260, Singapore Received 7 May 2004; in revised form 16 July 2004 Available online 20 August 2004
Abstract The infrared absorption spectrum of the m12 fundamental band of ethylene-d (C2H3D) has been recorded with an unapodized resolution of 0.004 cm1 in the wavenumber range of 1340–1460 cm1 using the Fourier transform technique. By assigning and fitting a total of 870 infrared transitions using a WatsonÕs A-reduced Hamiltonian in the Ir representation, three rotational and five quartic centrifugal distortion constants for the upper state (v12 = 1) were determined for the first time. The rms deviation of the fit was 0.00044 cm1 which is close to the experimental precision of the absorption lines. The A-type m12 band centred at 1400.762811 ± 0.000041 cm1was found to be relatively free from local frequency perturbations. The inertial defect D12 was found ˚ 2. to be 0.20928 ± 0.00002 lA 2004 Elsevier Inc. All rights reserved. Keywords: FTIR spectrum of ethylene-d
1. Introduction In the past decades, low-resolution infrared measurements [1–4] have been made on the ethylene-d (C2H3D) molecule and some of its isotopomers for the understanding of their basic vibrational structures. In the studies of Duncan et al. [1,4], most of the vibrational bands of C2H3D were identified and assigned accurately. Furthermore, infrared measurements and analyses [5] of C2H3D in the 2830–3270 cm1 region with a resolution better than 0.030 cm1 were conducted to study the Fermi resonance between m1 and m5, and to derive rotational and centrifugal distortion constants for m5, m2 + m3, and m9. The pure rotational spectra of C2H3D were observed and analysed by microwave spectroscopy [6]. In this
*
Corresponding author. Fax: +65 6896 9414. E-mail address: [email protected] (T.L. Tan).
0022-2852/$ - see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jms.2004.07.008
work [6], accurate rotational and centrifugal distortion constants in the ground state were determined, and the average structure of C2H3D molecule was also calculated. Most of the recent high resolution (at 0.004 cm1 or better) FTIR studies on the ethylene molecules were focussed on C2H4 [7,8], C2D4 [9–11], trans-C2H2D2 [12,13], cis-C2H2D2 [14,15], and 1,1-ethylene-d2 [16]. However, FTIR spectroscopic works on C2H3D at resolution of 0.004 cm1 or better have not been reported so far. The purpose of this work is to measure and analyse the m12 band of ethylene-d (C2H3D) which was measured with a resolution of 0.004 cm1. By fitting 870 assigned transitions, the three rotational and five quartic centrifugal distortion constants of the v12 = 1 state were obtained for the first time. The rms deviation of the fit of transitions of the A-type m12 band was 0.00044 cm1. As the rms deviation of the fit is close to the estimated absolute accuracy (0.0005 cm1) of the measured
106
T.L. Tan et al. / Journal of Molecular Spectroscopy 228 (2004) 105–109
transition, it can be concluded that the m12 band is relatively free of local frequency perturbations.
2. Experimental details The spectra of the m12 band of C2H3D were recorded using a Bomem DA3.002 Fourier transform spectrophotometer [8–11], with a unapodized resolution of 0.004 cm1. A Globar infrared source, and a high sensitivity liquid nitrogen cooled HgCdTe detector, and KBr beam splitter were used. A low-pass infrared band filter (0–2500 cm1) was used to reduce background noise level. The final spectrum was produced by coadding four runs of 60 scans each with the total scanning time of about 23 h. The linewidth (FWHM) in the final spectrum was found to be about 0.0045 cm1. The ethylene-d (C2H3D) gas samples of chemical purity better than 98%, used in these measurements were supplied by Cambridge Isotope Laboratories (Cambridge, MA). For the spectral measurements at the ambient temperature of about 296 K, we used a single pass 20-cm absorption cell with ZnSe windows. The vapor pressure of about 5 Torr in the cell was measured using capacitance pressure gauge. A spectrum of the evacuated cell, consisting of 200 scans at a resolution 0.1 cm1, was also measured and transformed at 0.004 cm1 with zero-filling to obtain a background spectrum. This spectrum when ratioed with the sample spectrum of C2H3D gave a transmittance spectrum with relatively smooth baseline. The spectrum in the 1330–1460 cm1 region showed some strong H2O absorption lines due to H2O impurity in the gas cell. However, the H2O lines did not cause any interference with those of C2H3D since the line positions of H2O are known accurately [17] and the lines are widely spaced. Calibration of the absorption lines of m12 of C2H3D was conducted using 34 unblended and unsaturated H2O lines in the 1330–1460 cm1 region. The selected H2O calibration frequencies were taken from Guelachvili and Rao [17]. A correction factor of 1.00000096 was required to bring the observed wavenumbers into agreement with the calibrated frequencies. From the line fitting the relative precision of the wavenumbers obtained was in the order of 0.0004 cm1. After considering small systematic errors in the experiments, it is reasonable to estimate the absolute accuracy of the measured C2H3D lines to be approximately ± 0.0005 cm1.
symmetry and is one of the 12 fundamental modes of vibrations [1,4]. The m12 mode is ascribed to an in-plane C–H bending. A plot of the spectrum of m12 recorded with a resolution of 0.5 cm1 in the 1330–1470 cm1 region is shown in Fig. 1. It shows the presence of a prominent strong central Q branch at about 1400.76 cm1, which is typical of an A-type band. In the rotational analysis, the band has been confirmed to be A-type. In the high resolution plot, strong absorption lines were easily observed with a regular spacing of about 1.66 cm1 which is about B + C. During the initial assignment of the P and R branches of the band, the strongest of these lines was assigned as J = Kc and Ka = 0 or 1 which were resolved with decreasing magnitude, up to J 0 = 30 and 27 in the P and R branches respectively. These transitions become unresolved doublets for higher J 0 ¼ K 0c and Ka = 0 or 1 values. Fig. 2 shows the asymmetry splitting for J 0 ¼ K 0c ¼ 4, Ka = 0 and 1 transitions in the J 0 = 4 cluster in the P branch. Asymmetry splitting was also observed for Ka = 2 in the J 0 = 4 cluster, while the transitions of Ka = 3, and 4 were not split and were unresolved doublets, as shown in Fig. 2. The intensity of the transitions in the same J 0 cluster decreases for higher Ka values. As the J 0 value increases the asymmetry splitting occurs for even higher Ka values for both P and R branches. Fig. 3 shows the asymmetry splitting for J 0 ¼ K 0c ¼ 5, Ka = 0and 1 transitions in the J 0 = 5 cluster in the R branch. Asymmetry splitting for Ka = 2 transitions is also observed. Since ground state rotational constants for C2H3D were available (6), initial assignment of the transitions could be accurately made, starting with low J values. Those low J transitions were fit to get preliminary constants that were used to predict the higher J transitions. These transitions were then assigned and the process repeated to extend the assignments to even higher quantum numbers. This bootstrap procedure allowed us to assign all the transitions accurately. Accurate rovibrational band constants up to five quartic terms obtained from the fitting of the P and R branch transitions were used to calculate the Q branch transitions with high precision. A total of 256 well-resolved Q branch transitions were confidently assigned and used in the final fit. The strongest lines were all doublets with two possible values of Kc. There were enough fairly unblended lines in the strong Q branch to ensure the accuracy of the present assignments. Fig. 4 shows the m12 Q branch region with well-defined K 0a clusters. For each K 0a cluster, the line intensity is maximum for J 0 ¼ K 0a and K 0c ¼ 0 or 1, and the intensity decreases as J 0 increases.
3. Assignments and frequency analysis 4. Results and discussion The C2H3D molecule is a simple asymmetric top planar molecule of Cs symmetry with asymmetry parameter j of about 0.89. The m12 band is infrared active with A 0
The nonlinear least-squares program which was previously used in the analyses of the m12 bands of 12C2H4
T.L. Tan et al. / Journal of Molecular Spectroscopy 228 (2004) 105–109
Fig. 1. A survey spectrum with resolution of 0.5 cm1 of the m12 band of C2H3D.
Fig. 2. A detailed section of the P branch region of C2H3D.
107
108
T.L. Tan et al. / Journal of Molecular Spectroscopy 228 (2004) 105–109
Fig. 3. A detailed section of the R branch region of C2H3D.
Fig. 4. A survey spectrum with resolution of 0.004 cm1 in the Q branch region of C2H3D.
T.L. Tan et al. / Journal of Molecular Spectroscopy 228 (2004) 105–109 Table 1 Ground state and upper state (v12 = 1) constants (cm1) for C2H3D (A-reduction, Ir representation)
A B C DJ · 106 DJK · 106 DK · 105 dJ · 107 dK · 106 m0 Number of infrared transitions Rms deviation (cm1) ˚ 2) D12 (lA
Ground state [6]a
v12 = 1
4.0058887(23) 0.91632515(87) 0.74377271(47) 1.3010(80) 6.012(66) 7.0599(53) 2.806(10) 8.114(79) —
4.0390015(12)b 0.92022114(63) 0.74255820(54) 1.36209(54) 4.6673(35) 8.05600(94) 3.0941(41) 9.503(24) 1400.762811(41) 870 0.000 44 0.209280(20)
0.059821(2)
a
All five quartic constants were included in the present fit. The uncertainty in the last digits (twice the estimated standard error) is given in parentheses. b
[8] and 12C2D4 [9] was also employed to fit the assigned transitions of m12 of C2H3D. This program for fitting asymmetric rotor spectra uses a Watson Hamiltonian [18] with an Ir representation in an A-reduction. Ground state rovibrational constants of C2H3D inclusive of all five quartic terms were initially taken from Hirota et al. [6]. These constants represent the most accurate values available for C2H3D. They were derived with high precision from 21 observed transitions in the microwave spectra of C2H3D. The accuracies of all five quartic constants were improved during the fit. In our rotational analysis, the ground state constants were fixed to determine the upper state v12 = 1 constants of C2H3D inclusive of all five quartic terms. Inclusion of sextic constants did not improve the accuracy of the fit. A total of 870 infrared transitions was finally assigned and fitted in the determination of the v12 = 1 rovibrational constants as given in Table 1. The m12 band centre is found to be 1400.762811 ± 0.000041 cm1. This value is smaller than that of 1401.5 cm1 measured by Duncan et al. [1], in 1973, in their study of 13C frequency shifts in ethylene. In a recent work by Duncan et al. [4], in 1993, the band centre was observed to be 1400.0 cm1. The fitting program, though written in terms of prolate matrix elements, had no problem in rapidly converging to the correct solution after several iterations. In the least square analysis, the infrared measurements were weighted by the inverse square of the estimated uncertainty. The infrared transitions were given an uncertainty of 0.0004 cm1 which is close to the absolute accuracy of the measured line. The rms deviation of the fit with all weighted lines was 0.00044 cm1. The transitions include values of Ka ranging from 0 to 12. For the P branch, we have J 0 and K 0c ranging from 0 to 30. For the R branch, J 0 and K 0c values range from 1 to 27. The values of J and Kc range from 1 to 22 and 0 to 13, respectively, for the Q branch. The wide ranging
109
values of J 0 , K 0a , and K 0c used in the analysis effectively cover the whole frequency range of 1340–1460 cm1 for the m12 band of C2H3D. ˚ 2 of C2H3D deThe inertial defect D12 of 0.20928(2) lA rived from the present analysis, as given in Table 1 is less ˚ 2; and that of than that of C2H4 [8] which is 0.24201(2) lA 2 ˚ trans-C2H2D2 [13] which is 0.29967(2) lA ; and also that ˚ 2. The upper of cis-C2H2D2 [15] which is 0.24820(2) lA state rovibrational constants A, B, DJ, DK, dJ, and dK of m12 band of C2H3D as given in Table 1 are found to be greater than the corresponding ground state values, while the constants C and DJK are lower than those of ground states. These trends in sign are in good agreement with those in the m12 band of C2H4 [8]. This result indicates similarity in the rotational structures of the m12 bands of both ethylene molecules. In conclusion, the present investigation reports the measurements, assignments, and fitting of 870 infrared transitions of the m12 band of C2H3D. Accurate upper state (v12 = 1) constants inclusive of all five quartic terms were derived for the first time. These spectral data would be another useful contribution to the understanding of the molecular structure of ethylene. References [1] J.L. Duncan, D.C. McKean, P.D. Mallinson, J. Mol. Spectrosc. 45 (1973) 221–246. [2] Y. Verbist-Scieur, C.P. Courtoy, A. Fayt, D. Van Lerberghe, Ann. Soc. Sci. Brux. 90 (1976) 317–336. [3] A. Fayt, Ann. Soc. Sci. Brux. 84 (1970) 69–106. [4] J.L. Duncan, A.M. Ferguson, S.T. Goodlad, Spectrochim. Acta 49A (1993) 149–160. [5] Y. Verbist-Scieur, C.P. Courtoy, A. Fayt, J. Mol. Spectrosc. 85 (1981) 480–492. [6] E. Hirota, Y. Endo, S. Saito, K. Yoshida, I. Yamaguchi, K. Machida, J. Mol. Spectrosc. 89 (1981) 223–231. [7] R. Georges, M. Bach, M. Herman, Mol. Phys. 90 (1997) 381–387. [8] T.L. Tan, S.Y. Lau, P.P. Ong, K.L. Goh, H.H. Teo, J. Mol. Spectrosc. 203 (2000) 310–313. [9] T.L. Tan, K.L. Goh, P.P. Ong, H.H. Teo, Chem. Phys. Lett. 315 (1999) 82–86. [10] K.L. Goh, T.L. Tan, P.P. Ong, H.H. Teo, Mol. Phys. 98 (2000) 583–587. [11] T.L. Tan, K.L. Goh, P.P. Ong, H.H. Teo, J. Mol. Spectrosc. 202 (2000) 249–252. [12] F. Hegelund, J. Mol. Spectrosc. 135 (1989) 45–58. [13] H.H. Teo, P.P. Ong, T.L. Tan, K.L. Goh, J. Mol. Spectrosc. 204 (2000) 145–147. [14] F. Hegelund, F.M. Nicolaisen, J. Mol. Spectrosc. 128 (1988) 321– 333. [15] K.L. Goh, T.L. Tan, P.P. Ong, H.H. Teo, Chem. Phys. Lett. 325 (2000) 584–588. [16] F. Hegelund, J. Mol. Spectrosc. 148 (1991) 415–426. [17] G. Guelachvili, K. Narahari Rao, Handbook of Infrared Standards, Academic Press, Orlando, FL, 1986. [18] J.K.G. Watson, in: J.R. Durig (Ed.), Vibrational Spectra and Structure, A Series of Advances, vol. 6, Elsevier, New York, 1977 Chapter 1.