Journal of Molecular Structure 695–696 (2004) 385–394 www.elsevier.com/locate/molstruc Rovibrational analysis of the n4 and n5 þ n9 bands of CHCl2F Sieghard Albert*, Karen Keppler Albert, Martin Quack* Laboratorium für Physikalische Chemie, ETH Zürich, Hönggerberg, CH-8093 Zürich, Switzerland Received 22 December 2003; accepted 22 December 2003 This paper is dedicated to Professor Dr Manfred Winnewisser and Dr Brenda Winnewisser on the occasion of his 70th and her 65th birthday Abstract The infrared spectrum of CHCl2F has been measured with a new, very high resolution Fourier transform infrared spectrometer, the Bruker IFS 120 HR Zürich Prototype (ZP) 2001. The spectrum was recorded with a resolution of 0.0007 cm21 in the range 600– 2300 cm21 at room temperature. The assignment of the rovibrational transitions has been carried out with the Giessen interactive Loomis – Wood program developed by Winnewisser et al. [J. Mol. Spectrosc. 136 (1989) 12] and the least squares adjustment has been performed with the Zürich 35 35 21 WANG program. The spectrum has been analyzed in the n4 region of CH Cl2F ðn~0 ¼ 744:474 cm Þ and the n5 þ n9 regions of CH Cl2F ðn~0 ¼ 829:084 cm21 Þ and CH35Cl37ClF ðn~0 ¼ 825:027 cm21 Þ using an effective Hamiltonian. Both bands are important to understand the absorption behavior of the fluorochlorohydrocarbon CHCl2F, important in the context of atmospheric pollution as well as in laser chemistry. Local perturbations have been identified in both bands. The results are discussed in relation to molecular parity violation in the case of the chiral isotopomer CH35Cl37ClF. q 2004 Elsevier B.V. All rights reserved. Keywords: High resolution spectroscopy; Resonance; Infrared; Atmospheric spectroscopy 1. Introduction The infrared spectra of fluorochlorohydrocarbon molecules and their rovibrational analysis are of crucial importance for understanding the absorption behavior of trace gases in the Earth’s atmosphere [2,3] (and references cited therein) as well as for understanding the infrared laser chemistry of these compounds [4,5] (and references therein). It turns out that for many of these compounds analyses were possible in the past only by using advanced FTIR and laser spectroscopic techniques in combination with supersonic jet cooling [6] (and references therein). Early analyses using these techniques were possible for CHF2Cl [7,8], CHFCl2 [3] and CFCl3 [9]. However, there has also been a considerable improvement of resolving power of commercial FTIR spectrometers over the last two decades. A first breakthrough arrived with the Bomem instruments achieving a resolution of about 0.004 cm21 (FWHM, apodized), which led to the early developments in high resolution FTIR-supersonic jet techniques [10 – 14]. * Corresponding authors. Tel.: þ 41-1-6324421; fax: þ 41-1-6321021. E-mail addresses: quack@ir.phys.chem.ethz.ch, martin@quack.ch (M. Quack), albert@ir.phys.chem.ethz.ch (S. Albert). 0022-2860/$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.molstruc.2003.12.047 Another important step was the development of the Bruker IFS 120 HR instrument with resolutions up to 0.0013 cm21 [15]. The past recent advance was achieved with the Bruker Zürich prototype 2001 (IFS120 HR ZP 2001), which achieves an instrumental resolution of about 0.0007 cm21 [16,17]. While the steps in these two later developments may appear small, at first sight, it turns out that they are very important because they make possible analyses of the spectra taken at room temperature for the type of molecules discussed here, avoiding the need for the complicated jet cooling experiments. Of course, for even more complex molecules, cooling techniques in conjunction with the new spectrometer may become relevant again, not only because they simplify the spectra recorded, but also because of their potential in reducing Doppler widths. We have recorded and analyzed here the FTIR spectrum of CHCl2F at room temperature in the range of 600– 2300 cm21 [18] at essentially Doppler-limited resolution. The rovibrational spectra of CH35Cl2F and CH35Cl37ClF have been already studied by Snels and Quack [3] for the n3 ; n8 and n7 fundamental bands. However, the resolution of the analyzed spectra in that work ranged from 0.008 to 0.004 cm21. As a result, the root mean square deviation of 386 S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 Table 1 Summary of the fundamentals of CHCl2F (the data for v1 and v2 are from an effective-Resonance Hamiltonian fit [39]) Vibration Symmetry for C5 isotopomers n1 n2 n3 n4 n5 n6 n7 n8 n9 a0 a0 a0 a0 a0 a0 a00 a00 a00 CH stretching CH bending CF stretching CCl stretching CFCl bending ClCCl bending CH bending CCl stretching CFCl bending the assigned lines was only drms ¼ 0:0015 – 0:0026 cm21 ; which was slightly smaller than the error associated with the determination of line positions from the spectra. The ground state of CHCl2F has already been analyzed using microwave [19] and recently, submillimeter wave spectroscopy [20]. Quadrupole coupling constants in the ground state have been determined [20,21]. Table 1 summarizes the current knowledge of fundamental vibrations of CHCl2F, including also the results of the present work. In this paper, we present the rovibrational analysis of two further bands of CHCl2F; the n4 fundamental of CH35Cl2F (CCl stretch, n~0 ¼ 744:47 cm21 ) and the n5 þ n9 combination bands of CH 35Cl 2F ðn~0 ¼ 829:08 cm21 Þ and CH35Cl37ClF ðn~0 ¼ 825:03 cm21 Þ which are important for atmospheric absorption [2]. The analysis of an additional spectral region 900 –2300 cm21 of CHCl2F, in particular of the 1800 – 2200 cm21 region (the overtone of the CF stretching mode, 2n3 ) will be presented in separate publications [22]. The chiral isotopomer CH35Cl37ClF is of potential importance for general aspects of isotopic chirality and parity violation [23,24]. 2. Experimental The FTIR spectrum of CHCl2F has been recorded in the region 600 – 2300 cm 21. The resolution, defined by 1/MOPD (maximum optical path difference) was 0.001 cm21. One hundred and fifty spectra were co-added in each spectral region. A White-type cell with a path length ranging from 3.2 up to 19.2 m was used. The sample pressure was varied from n~ (cm21) Reference CH35Cl2F CH35Cl37ClF 3024.8 1312 1079.4276 744.474290 458.2 277.2 1239.1940 807.1749 367.4 3024.8 1312 1079.3893 741.4 455.8 274.3 1238.6884 805.0492 365.1 [39] [39] [3] this work, [40] [40] [40] [3],[39] [3] [40] 0.09 to 1.3 mbar. An absorption path length of 19.6 m and a pressure p ¼ 1:3 mbar were used for the n4 band. For the n5 þ n9 band, the path length was 2.3 m and the pressure p ¼ 0:5 mbar: Pressure broadening was unimportant under these conditions. All spectra were self-apodized. Table 2 shows the measurement parameters. All spectra were recorded essentially at Doppler-limitation (Dn~Doppler ¼ 0:0012 cm21 at 1000 cm21 and 295 K). Apertures of 0.8, 1.0 and 1.15 mm were used, which led to instrumental resolutions of up to 0.0007 cm21. The spectra were calibrated with OCS [25]. 3. Assignment CHCl2F exists in three isotopomers with a natural abundance of approximately 9:6:1 (CH 35Cl 2F: CH35Cl37ClF: CH37Cl2F). Due to the two heavy atoms and the presence of several isotopomers, the rotational structure of the bands is dense and congested. Only hybrid bands have been observed in the spectrum. In this work, the n4 band of CH35Cl2F and the n5 þ n9 bands of the two major isotopomers CH35Cl2F and CH35Cl37ClF have been analyzed. In spite of the congested spectrum at room temperature it was not necessary to decrease the rotational temperature of the sample in order to simplify the spectra because the different series of transitions were clearly visible in the spectra. In the following description, the term ‘subband’ or ‘series’ describes the transitions belonging to one Ka or Kc value. The assignment of the observed rovibrational transitions belonging to a particular subband consisting of P and R Table 2 Experimental setup for the region 600–1000 cm21 of the infrared spectrum of CHCl2F Region (cm21) Resolution (cm21) Windows Source Detector Beamsplitter Opt. filter (cm21) Aperture (mm) nmirror (kHz) Electr. filter (cm21) Calib. gas 600 –1000 0.0007 KBr Globar MCT Ge:KBr 550–1000 0.8, 1.0, 1.15 40 395–1580 OCS [25] S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 branches has been carried out efficiently with an interactive Loomis –Wood assignment program [1] developed in the Giessen Laboratory. This method, previously extensively used for the assignment of linear molecules [26], has been used successfully for several asymmetric top molecules (CH35Cl2F, CDBrClF [17] and C2H3DO [27]) because these molecules display near symmetric top behavior (no observable asymmetry splitting) at higher J levels. 3.1. The n4 (CCl stretch) band of CH35Cl2F The n4 band is the weakest band of the nine fundamental bands of CHCl2F. A sufficient signal-to-noise ratio was only obtained for the main species CH35Cl2F. a-type transitions up to J # 91 and c-type transitions up to J # 58 have been assigned in the spectrum. The a-type series have been identified as P and R branches with J ^ 1; Ka ; Kc ¼ J ^ 1 2 Ka ˆ J; Ka ; Kc ¼ J 2 Ka : The spacing between two transitions of a series is approximately 2C: The assignments have been checked by comparison of the combination differences (CDs) of the ground state calculated from the assignments of the n3 ; n4 ; n5 þ n9 ; n7 and n8 fundamental bands. As the Loomis – Wood diagram of the a-type region of the n4 band of CH35Cl2F in Fig. 1 shows, there are numerous local resonances. At least three crossings, at J ¼ 35=39; J ¼ 41=44 and J ¼ 62=67; have been detected. The perturbations increase with increasing J and Ka quantum number. The n5 þ n6 and 2n9 levels are the most likely partners in these interactions with n4. The c-type subbands shown in Fig. 2 consist also of P and R branches with J ^ 1; Ka ¼ J ^ 1 2 Kc ; Kc ˆ J; Ka ¼ J 2 Kc ; Kc : The spacing between two transitions of 387 a series is approximately 2A: Perturbations have not been detected. As one can see from Figs. 2 and 3, all the ctype R branches are grouped together into three groups. In contrast, the c-type P branch series are grouped together into two groups. It is interesting to note that the c-type transitions have been assigned only as P and R branch transitions in the n4 and n3 bands. In all the other bands analyzed, n8 ; n7 [18] and n5 þ n9 ; the c-type transitions have been detected as Q branch transitions. 3.2. The n5 þ n9 combination band of CH35Cl2F and CH35Cl37ClF The signal-to-noise ratio was sufficient to assign the bands of the two major isotopomers CH35Cl2F and CH35Cl37ClF. The a-type transitions (J ^ 1; Ka ; Kc ¼ J ^ 1 2 Ka ˆ J; Ka ; Kc ¼ J 2 Ka Þ have been assigned in P and R branches up to J # 118 for CH35Cl2F, and J # 73 for CH35Cl37ClF whereas the c-type transitions (J; Ka ¼ J 2 Kc ; Kc ˆ J; Ka ¼ J 2 Kc ; Kc ) have been identified as Q branches up to J ¼ 55 for CH35Cl2F and J ¼ 40 for CH35Cl37ClF. The Loomis – Wood diagrams in Fig. 4 show a-type transitions of the two major isotopomers CH35Cl2F (straight series) and CH35Cl37ClF (slanted series). As can be seen in Fig. 4, the a-type transitions of CH35Cl2F are locally perturbed at J ¼ 59=67: The 2n6 þ n9 state probably acts as the resonance partner. Within the observed range of a-type transitions of CH35Cl37ClF, no crossings have been detected. However, the lines have been only assigned up to J ¼ 73: Fig. 1. Loomis–Wood diagram of the a-type transitions of the n4 band of CH35Cl2F in the R branch region. Local perturbations indicated by arrows are visible. 388 S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 Fig. 2. Loomis– Wood diagram of the c-type transitions of the n4 band of CH35Cl2F within the R branch region. The transition series are grouped into three groups. The spectrum in the area of the outlined part of the Loomis–Wood plot between 762.00 and 762.35 cm21 is shown in Fig. 3. 4. Discussion The rovibrational analysis has been carried out with Watson’s S and A reduced effective Hamiltonians in the I r representation up to sextic centrifugal distortion constants [28] in order to test the different models. The spectroscopic data have been analyzed with the WANG program described in detail in Ref. [8]. For the A reduction, the spectroscopic constants of the ground state constants were fixed to the values of Luis et al. [20] for CH35Cl2F and CH35Cl37ClF during the fits. For the S reduction, no ground state values were available. For that reason, Fig. 3. A comparison of measured R branch transitions (top trace, T ¼ 295 K; path length ¼ 19.6 m, Doppler-limited resolution ¼ 0.0012 cm21) of the n4 band of CH35Cl2F with a simulation (lower trace, resolution ¼ 0.0012 cm21). This is the part of the R branch where the transition series within one group fall together (see Fig. 2). The noise in the experimental spectrum is visible. 389 S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 Fig. 4. Loomis–Wood diagram of the a-type transitions of the n5 þ n9 band of CH35Cl2F. Local perturbations of the different Ka series are visible and indicated by circles. The slanted series belong to CH35Cl37ClF. the constants of the ground states of CH35Cl2F and CH35Cl37ClF have been determined by fitting the CDs calculated from the assignments of the n3 ; n4 ; n7 ; n8 and n5 þ n8 bands and the rotational transitions from McLay [19]. The constants are listed in Table 3 for CH35Cl2F and Table 4 for CH35Cl37ClF. The deviations of the values of rotational constants of the ground state fitted by Watson’s S and A reduced Hamiltonians are only visible in the sixth digit after the decimal point. For the calculation of the CD, J levels up to J # 90 and Ka # 12 within the a-type transitions of CH35Cl2F Table 3 Spectroscopic constants in cm21 of the ground state and the n4 and n5 þ n9 states of CH35Cl2F in the S and A reductions S n~0 A B C DJ £ 106 DJK £ 106 DK £ 106 d1 £ 106 d2 £ 106 HJ £ 1012 HJK £ 1012 HKJ £ 1012 HK £ 1012 h1 £ 1012 h2 £ 1012 h3 £ 1012 N drms Jmax for fit Jmax assigned Ka max a-types Kc max c-types A n~0 A B C DJ £ 106 DJK £ 106 DK £ 106 dJ £ 106 dK £ 106 FJ £ 1012 FJK £ 1012 FKJ £ 1012 FK £ 1012 fJ £ 1012 fJK £ 1012 fK £ 1012 n4 Ground state n5 þ n9 S A [20] S A S A 0 0.233 118 37 (91) 0.110 318 6 (37) 0.078 385 11 (65) 0.114 39 (65) 20.118 6 (20) 0.014 7 (14) 20.058 70 (44) 20.027 25 (15) 0.082 4 (78) – – 20.092 0 (89) 0.209 (32) 0.150 (29) 0.064 2 (15) 0 0.233 120 688 0.110 316 968 0.078 386 106 0.030 914 1 20.030 286 0.286 28 0.010 091 5 0.057 293 0 0.031 054 8 20.171 11 20.266 85 1.330 92 – 0.306 89 20.540 374 744.474 290 (91) 0.232 929 58 (37) 0.109 705 8 (11) 0.078 102 24 (27) 0.114 63 (89) 20.123 7 (28) 0.018 4 (20) 20.058 38 (48) 20.024 76 (15) 0.082 4 – – 20.092 0 0.261 (64) 0.150 (66) 0.101 (49) 744.474 354 (71) 0.232 931 66 (17) 0.109 704 30 (58) 0.078 103 09 (23) 0.027 72 (34) 20.009 51 (88) 0.262 9 (10) 20.009 16 (17) 0.062 9 (23) 0.031 054 8 20.171 11 20.266 85 1.330 92 – 0.306 89 20.540 374 829.084 193 (46) 0.232 457 21 (25) 0.110 317 49 (30) 0.078 365 47 (13) 0.112 99 (13) 20.118 10 (37) 0.014 71 (26) 20.055 92 (14) 20.028 444 (74) 20.083 7 (87) – – 20.092 0 20.280 (33) 0.558 (30) 0.064 2 829.084 205 (47) 0.232 459 60 (26) 0.110 315 13 (27) 0.078 366 58 (13) 0.033 38 (20) 20.047 13 (90) 0.296 02 (68) 0.011 70 (10) 0.059 72 (70) 20.077 1 (87) 20.171 11 20.266 85 1.150 (61) – 0.306 89 232.3 (22) 850 0.000 28 90 105 12 27 206 – 80 – – – 0.000 34 1230 0.000 34 80 91 7 15 0.000 27 1680 0.000 28 80 118 12 10 If there are no uncertainties listed, the constant was fixed during the fit. The uncertainties given in parentheses one standard deviation in units of the last digits given. 390 S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 Table 4 Spectroscopic constants in cm21 of the ground state and n5 þ n9 state of CH35Cl37ClF in the S and A reductions A S n~0 A B C DJ £ 106 DJK £ 106 DK £ 106 d1 £ 106 d2 £ 106 HJ £ 1012 HJK £ 1012 HKJ £ 1012 HK £ 1012 h1 £ 1012 h2 £ 1012 h3 £ 1012 N drms Jmax for fit Jmax assigned Ka max a-types Kc max c-types n~0 A A A DJ £ 106 DJK £ 106 DK £ 106 dJ £ 106 dK £ 106 FJ £ 1012 FJK £ 1012 FKJ £ 1012 FK £ 1012 fJ £ 1012 fJK £ 1012 fK £ 1012 n5 þ n9 Ground state S A [20] S A 0 0.231 608 9 (14) 0.107 395 6 (44) 0.076 734 01 (76) 0.119 34 (20) 20.145 7(68) 0.037 0(48) 20.055 5(10) 20.027 9(25) 20.228(57) – – 20.295(49) 0.687(12) 0.027 5(64) 0.372(69) 0 0.231 607 360 0.107 392 039 0.076 735 753 0.029 401 0.029 350 0.283 840 0.009 538 1 0.055 748 6 0.031 215 20.160 778 20.286 532 1.330 92 – 0.306 89 20.540 374 825.027 526 (51) 0.230 951 60 (62) 0.107 385 47 (33) 0.076 699 149 (93) 0.111 50 (48) 20.133 07(61) 0.031 92(21) 20.053 08(30) 20.031 57(16) 20.235 3(61) – – 20.295 0.569(27) 0.027 5 0.372 825.027 505 (35) 0.230 950 68 (56) 0.107 382 30 (17) 0.076 700 761 (37) 0.031 029 (86) 20.005 09(75) 0.312 3(13) 0.010 509(43) 0.049 40(42) 0.031 215 20.160 778 20.286 532 1.330 92 – 0.306 89 20.540 374 580 0.000 31 67 83 8 17 125 – 60 – – – 1080 0.000 28 0.000 28 66 73 9 10 If there are no uncertainties listed, the constant was fixed during the fit. (See also table 3). and J # 65 and Kc # 27 within the c-type transitions of CH35Cl2F were used. For CH35Cl37ClF, J # 65 and Ka # 8 within the a-type transitions and J # 55 and Kc # 17 within the c-type transitions were used. The sextic constants HJK and HKJ of Watson’s S reduced Hamiltonian could not be determined and were set to zero for CH35Cl2F and CH35Cl37ClF. In total 850 CD of CH35Cl2F and 580 CD of CH35Cl37ClF were used. The standard deviation is slightly better for the CH35Cl2F isotopomer than for the CH35Cl37ClF isotopomer because the signal-to-noise ratio of the CH35Cl2F absorption lines is higher. Based on these fixed ground state constants, the constants of the n4 level of CH35Cl2F and the n5 þ n9 levels of CH35Cl2F and CH35Cl37ClF were determined. These are shown in Tables 3 and 4. Local resonances of the involved bands have been neglected during the fit by not including transitions in perturbed regions. The sextic constants HJK and HKJ of Watson’s S reduced Hamiltonian of the n4 level were set to zero as was also done for the ground state. In addition, HJ and HK were fixed to the values of the ground state. The sextic constants could not be determined for the A reduced Hamiltonian of the n4 level and were fixed to the values of the ground state. The standard deviations drms for the fits of the n4 level are identical for both models (S and A reduced Hamiltonians). The spectrum of the n4 band for CH35Cl2F was simulated based on these constants and was identical for both reductions. A comparison of the measured and simulated c-type R branch lines is shown in Fig. 3. This is the part of the R branch region in which the series within the three groups fall together. The same part of the spectrum (762.00 – 762.35 cm21) is outlined in the Loomis – Wood diagram in Fig. 2. The agreement between simulation and experiment in Fig. 3 is quite good. It is evident that there are more lines in the measured spectrum than in the simulation. These extra lines belong probably to the CH35Cl37ClF isotopomer. However, the relatively high noise level of the recorded spectrum is clearly visible. This makes it practically impossible to recognize series of absorption lines of the CH35Cl37ClF isotopomer in this region. A comparison of the change of the rotational constant B upon excitation of the modes n3 ; n4 ; n7 ; n8 and n5 þ n9 [3,18] shows the largest change upon excitation of n4 (Table 3). The sextic constants HJK and HKJ of the S reduction for the n5 þ n9 levels of CH35Cl2F and CH35Cl37ClF could also not be determined and were set to zero. In addition, the HK and h3 constants for CH35Cl2F and HK and h2 and h3 for CH35Cl37ClF were fixed to the values for the ground state. As the standard deviations show for the two isotopomers, the anharmonic resonance mentioned in the paper of Snels and Quack [3] can be neglected for a line-by-line assignment of individual bands, providing effective parameters only. However, the large value of fK of CH35Cl2F for the A reduction is an indicator of a hidden global resonance. For CH 35Cl 2F the sextic constants FJ ; FK and fK of S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 391 Fig. 5. A comparison of the measured Q branches (c-type transitions) of the n5 þ n9 band of CH35Cl2F (top trace, T ¼ 295 K; path length ¼ 19.6 m, p ¼ 0:5 mbar; Doppler-limited resolution ¼ 0.0012 cm21) with a simulation of CH35Cl2F and CH35Cl37ClF (lower trace, resolution ¼ 0.0012 cm21). ‘the A reduction have been determined. For CH35Cl37ClF, no sextic constants in the A reduction could be determined. All the constants were fixed to the values of the ground state. The standard deviations drms for the both reduced Hamiltonians are again identical for both isotopomers. The simulation of the n5 þ n9 band based on the fitted constants, again identical for both models, illustrates an excellent agreement with the measured spectrum. Fig. 5 shows a comparison of measured and simulated c-type Q branch lines of CH35Cl2F up to J ¼ 54: A part of the Q branch is enlarged in Fig. 6 to demonstrate the resolution of the new Zurich spectrometer. Even for the weaker lines, atype R branch lines of CH35Cl37ClF and a-type P branch lines of CH35Cl2F at low J values, the agreement between Fig. 6. A comparison of an expanded part of the measured Q branch lines of the n5 þ n9 band of CH35Cl2F (top trace, T ¼ 295 K; path length ¼ 19.6 m, p ¼ 0:5 mbar; Doppler-limited resolution ¼ 0.0012 cm21) with a simulation of CH35Cl2F and CH35Cl37ClF (lower trace, resolution ¼ 0.0012 cm21). Parts of the a-type P branches of CH35Cl2F and R branches of CH35Cl37ClF are also visible. 392 S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 Fig. 7. A comparison of the measured Q branches of CH35Cl37ClF and P branches of CH35Cl2F of the n5 þ n9 band (top trace, T ¼ 295 K; path length ¼ 19.6 m, p ¼ 0:09 mbar; Doppler-limited resolution ¼ 0.0012 cm21) with a simulation of CH35Cl37ClF and CH35Cl2F (lower trace, resolution ¼ 0.0012 cm21). recorded and simulated spectrum is superb. Another example is shown in Fig. 7. Due to the high resolving power of the spectrometer we can partially resolve Q branch features. In addition to c-type transitions of Q branches of CH35Cl37ClF there are strong a-type P branches of CH35Cl2F. A part of the a-type P branches of CH35Cl2F is shown in Fig. 8. In spite of a fairly high noise level, the agreement between measurement and simulation is very Fig. 8. A comparison of a part of the measured P branches of the n5 þ n9 band of CH35Cl2F (top trace, T ¼ 295 K; p ¼ 1:3 mbar; path length ¼ 19.6 m, Doppler-limited resolution ¼ 0.0012 cm21) with a simulation of CH35Cl2F and CH35Cl37ClF (lower trace, resolution ¼ 0.0012 cm21). S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 good. The patterns of the a-type transitions are clearly visible. 5. Conclusions The recording and assignment of congested and complicated spectra of relatively heavy molecules ðm $ 100 DaÞ is much simpler with the higher resolution available with our new spectrometer because we can avoid complicated supersonic jet cooling techniques. The spectra can be recorded at room temperature, which saves a great deal of time and effort. The Giessen Loomis – Wood assignment program [1] is very helpful in assigning the spectra of prolate and oblate asymmetric rotors and the WANG program [8] provides a fairly universal tool in fitting effective Hamiltonian constants to complex band systems. Watson’s reduced A and S Hamiltonians both reproduce the recorded spectra very well for the bands analyzed here. We have analyzed for the first time the rovibrational structure of the n4 band of CH35Cl2F and the combination band n5 þ n9 of CH35Cl2F and CH35Cl37ClF. Both bands are of importance for a complete understanding of the absorption behavior of the fluorochlorohydrocarbon CHCl2F in the Earth’s atmosphere. Due to the high resolving power of our spectrometer, we are now able to detect local resonances which could not be observed in earlier studies [3]. The assignment of the bands showed that the a-type transitions of the n4 and n5 þ n9 bands of CH35Cl2F are locally perturbed. Interesting resonances were found in the previously assigned bands n3 and n7 of CH35Cl2F. In particular, the n3 band illustrates perturbations in the a- and c-type transitions. The analysis of these bands and the 2n3 band will be the subject of a future paper [22]. Bands of the chiral isotopomer CH35Cl37ClF which fall within the CO2 laser region are of interest for laser chemistry [4,5] and for a future detection of molecular parity violation. The present analysis makes it possible to identify unperturbed regions in the spectrum of CH35Cl37ClF in which CO2 laser spectroscopy can be conducted. In particular, the 2n3 band is of interest because quasiresonantly enhanced two photon CO2 laser transitions can be used for this band. With the previous ground work on isotopic chirality in general, presented in Refs. [23,24,29] and here for the isotopomer CH35Cl37ClF, in particular, it might be possible to detect rovibrational absorption line shifts due to parity violation [30,31]. Another possibility to detect rotational line shifts is presented by submillimeter wave spectroscopy. Backward wave oscillators [32,33] have spectral purities smaller than 1 Hz. If they are phase locked they reach a resolving power of Dn=n ¼ 3 £ 10213 [32]. This ratio is probably not good enough for the chiral isotopomer CH35Cl37ClF, but it may be sufficient for heavier molecules [23,24,34,35]. 393 Another more obvious application of the present results concerns atmospheric spectroscopy. The importance of detailed rovibrational analyses for a correct interpretation of remote sensing experiments as well as for quantitative detection of tropospheric freons has been emphasized repeatedly [36– 38]. In the aftermath of the Montreal protocol banning the fluorochlorocarbons, the fluorochlorohydrocarbons may become more important in the future. These are considered less problematic because of their shorter lifetimes, which are due to kinetic decay by chemical attack at the H-atom. In this context, we are also working on a new determination of the absolute band strengths of the most important bands [22]. Acknowledgements Our work is financially supported by the Schweizerische Nationalfonds and the ETH Zürich (including AGS, CSCS and C4). We enjoyed discussions with Michael Hippler, Hans Hollenstein, Achim Sieben, Jürgen Stohner, Andreas Steinlin and Martin Willeke. References [1] B.P. Winnewisser, J. Reinstädtler, K.M.T. Yamada, J. Behrend, J. Mol. Spectrosc. 136 (1989) 12–18. [2] C.P. Rinsland, D.W. Johnson, A. Goldman, J.S. Levine, Nature 337 (1985) 535. [3] M. Snels, M. Quack, J. Chem. Phys. 95 (1991) 6355–6361. [4] D.W. Lupo, M. Quack, Chem. Rev. 87 (1987) 181–216. [5] M. Quack, Infrared Phys. 29 (1989) 441– 466. [6] M. Quack, Annu. Rev. Phys. Chem. 41 (1990) 839– 874. [7] A.J. Ross, A. Amrein, D. Luckhaus, M. Quack, Mol. Phys. 66 (1989) 1273– 1277. [8] D. Luckhaus, M. Quack, Mol. Phys. 68 (1989) 745 –758. [9] M. Snels, A. Beil, H. Hollenstein, M. Quack, U. Schmitt, F. D’Amato, J. Chem. Phys. 103 (1995) 8846–8853. [10] H.-R. Dübal, M. Quack, U. Schmitt, Chimia 38 (1984) 438– 439. [11] A. Amrein, M. Quack, U. Schmitt, Mol. Phys. 60 (1987) 237–248. [12] A. Amrein, M. Quack, U. Schmitt, Z. Phys. Chem. N.F. 154 (1987) 59–72. [13] A. Amrein, M. Quack, U. Schmitt, J. Phys. Chem. 92 (1988) 5455– 5466. [14] A. Amrein, D. Luckhaus, F. Merkt, M. Quack, Chem. Phys. Lett. 152 (1988) 275–280. [15] M. Birk, M. Winnewisser, E.A. Cohen, J. Mol. Spectrosc. 136 (1989) 420–443. [16] S. Albert, M. Quack, Chimia 56 (2002) 374. [17] S. Albert, K.K. Albert, M. Quack, FTS OSA Tech. Digest (2003) 177. [18] S. Albert, M. Quack, In Eighteenth Colloquium on High Resolution Molecular Spectroscopy, Paper D37, 2003, p. 145 Dijou France. [19] D.B. McLay, Can. J. Phys. 42 (1964) 720. [20] A. Luis, J.C. Lopez, A. Guarnieri, J.L. Alonso, J. Mol. Struct. 413 (1997) 249–253. [21] J.C. Lopez, A. Luis, S. Blanco, A. Lesarri, J.L. Alonso, J. Mol. Struct. 612 (2002) 287 –303. [22] S. Albert, M. Quack, A. Steinlin, in preparation. [23] M. Quack, Angew. Chem. Int. Ed. (Engl.) 28 (1989) 571–586. [24] M. Quack, Angew. Chem. Int. Ed. (Engl.) 114 (2002) 4812–4825. 394 S. Albert et al. / Journal of Molecular Structure 695 –696 (2004) 385–394 [25] A.G. Maki, S. Wells, Wavenumber calibration tables from heterodyne frequency measurements, NIST Special Publication 821, USA, 1991. [26] S. Albert, M. Winnewisser, B.P. Winnewissser, Ber. Bunsenges. Phys. Chem. 100 (1996) 1876–1896. [27] K.K. Albert, S. Albert, M. Quack, J. Stohner, O. Trapp, V. Schurig, In Eighteenth Colloquium on High Resolution Molecular Spectroscopy, Paper F38, 2003, p. 202 Dijou, France. [28] J.K.G. Watson, in: J.R. Durig (Ed.), Vibrational Spectra and Structure, vol. 6, Elsevier, Amsterdam, 1978, pp. 1–89. [29] R. Berger, M. Quack, A. Sieben, M. Willeke, in preparation. [30] M. Quack, J. Stohner, Phys. Rev. Lett. 84 (2000) 3807–3810. [31] M. Quack, J. Stohner, J. Chem. Phys. 119 (2003) 11228–11240. [32] F. Lewen, R. Gendriesch, I. Pak, D. Paveliev, R. Hepp, M. Schieder, G. Winnewisser, Rev. Sci. Instrum. 69 (1998) 32. [33] D.T. Petkie, T.M. Goyette, R.P.A. Bettens, S.P. Belov, S. Albert, P. Helminger, F.C. De Lucia, Rev. Sci. Instrum. 68 (1997) 1675– 1683. [34] R. Schwerdtfeger, J. Gierlich, T. Bollwein, Angew. Chem. Int. Ed. 42 (2003) 1293–1296. [35] M. Gottselig, M. Quack, J. Stohner, M. Willeke, Int. J. Mass. Spectrum (2004) in press. [36] N. Van-Thank, I. Rossi, A. Jean-Louis, H. Rippel, J. Geophys. Res. D. Atmos. 91 (1986) 4056–4062. [37] D.P. Kratz, P. Varanasi, J. Quant. Spectrosc. Radiat. Transfer 48 (1992) 245 –254. [38] S.T. Massie, P. Goldman, J. Quant. Spectrosc. Radiat. Transfer 48 (1992) 713 –719. [39] H.-R. Dübal, M. Quack, Mol. Phys. 53 (1984) 257–264. [40] A. Baldacci, A. Passerini, S. Ghersetti, Spectrochim. Acta, Part A 40 (1984) 165 –171.