Abstract
Two infrared absorption bands of CH2D2 have been analyzed in the semirigid rotor approximation. These are the A-type band at 2671.67 cm−1 and the C-type band at 4425.61 cm−1. The A-type band has previously been assigned as v3+v9, and the C-type band is tentatively assigned as v3+v6 The upper state of the A-type band is perturbed presumably by the close lying level 2v5. This interaction has not been investigated. The following values were found for the rotational constants of the ground vibrational state: A0=4.303 cm−1, B0= 3.504 cm−1, C0= 3.049 cm−1.
1. Introduction
The rotational-vibrational spectra of all the deuterated species of methane except CH2D2 have been well investigated [1].1
Methane and CD4 are spherical tops while CH3D and CD3H are symmetric rotors; on the other hand CH2D2 is an asymmetric rotor. It seemed of interest to determine the rotational constants of CH2D2 in order to have a set of constants for each of the species. Fortunately, high resolution spectra could be obtained for both an A-type and a C-type band for this molecule. As recently discussed [2], this is sufficient data to enable a good determination of the ground state constants using the complementary values obtained from the two bands. These two bands have been analyzed in the semirigid approximation to yield good values of the ground state constants. Unfortunately a perturbation in the excited state of the A-type band introduces an uncertainty in the effective constants for this band that is larger than can be justified for the precision of the data. Only transitions involving levels with low values of the rotational quantum numbers have been used in the analysis in order to minimize the effect of centrifugal distortion.
2. Experimental Procedure
The spectra were recorded with the grating instrument of the Infrared Spectroscopy Section [3] using a 10,000 lines/in. grating with a ruled area of about 6×8 in. A cooled PbS photoconductive cell was used as the detector.
Both the A- and C-type bands were recorded using a multiple reflection cell with a total path length of 6 m, filled with CH2D2 to a pressure of 2 cm of Hg. The C-type band was observed in the second order of the grating and the A-type band in the first. The CH2D2 obtained from Merck & Co., Ltd., had a stated minimum purity of 98%.
The wavelengths of the lines were measured using higher order infrared emission lines of the rare gases as standards, and interpolating between them through the use of the fringes of a Fabry-Perot interferometer as previously described [4].
3. Theory
Preliminary to the actual analysis of the spectra the mean square values of the angular momenta about the three principal axes of inertia and the intensities were calculated.
Assuming tetrahedral geometry and identical bond lengths of 1.094 A for CH2D2 the moments of inertia, reciprocal moments in cm−1 units, and the asymmetry parameter κ were obtained.
For the calculated value of κ=−0.27, α, β, and γ, as defined by Allen [2], were obtained by linear interpolation in published tables [5] of E(κ) for each energy level.
α, β, and γ may be shown to be identical with , , and , respectively, in an Ir representation [6]. The energy of a rotational level in a given vibrational state, neglecting centrifugal distortion, may be written as [2]:
and the difference between two rotational levels in the same vibrational state as:
The Δα’s, Δβ’s and Δγ’s for each of the ΔF2’s observable from strong transitions in the A- and C-type band were calculated.
The strong transitions for the A-type band are those which satisfy the selection rules:
and for the C-type band the strong transitions are those for which
The relative intensities of the transitions were calculated by combining the Boltzmann factor, calculated using the estimated moments of inertia, and nuclear spin statistics with the appropriate line strengths from published tables [7].
The nuclear spins of the equivalent pairs of hydrogen and deuterium atoms in CH2D2 give rise to degeneracies of the rotational levels. For the ground vibrational state the statistical weight factors are 15 for the symmetric (A) rotational levels and 21 for the antisymmetric (B) levels.
The A-type band at 2672 cm−1 has been previously observed by Wilmshurst and Bernstein [8] who assigned it to the combination v3+v9. As the observed band type is consistent with this assignment and there should be no other bands of this type near 2670 cm−1, there appears to be no reason to doubt the assignment.
The C-type band at 4425 cm−1 can best be assigned at v3+v6 as this seems to be the only combination which would give a C-type band in the region 4400–4500 cm−1.
4. Analysis
As each of the bands was isolated from others of comparable intensity, initial assignments, with the aid of the calculated intensities, could be made by inspection. From the initial assignments ΔF2 values were obtained, and with the calculated Δα’s, Δβ’s and Δγ’s were used to refine the values of A, B, and C in the ground and upper vibrational states. From these values of A, B, and C, α’s, β’s, and γ’s pertinent to the value of κ in each vibrational state were calculated as before.
A trial spectrum was then calculated from the expression:
From the trial spectrum and intensities more transitions could be assigned, enabling further refinement of the reciprocal moments of inertia.
While this iterative procedure worked well with the C-type band, Coriolis perturbations in the A-type band caused some difficulty in definitely locating some of the transitions.
The lowest observably perturbed level in the v3+v9 vibrational state is the 413 level which is pushed down by 0.24 cm−1. For J=5, the levels 505, 514, and 523 are all perturbed, and for J= 6 over half of the levels are perturbed.
This perturbation has not been investigated in detail. It probably arises through interaction with the vibrational state 2v5, the fundamental of which is theoretically inactive in the infrared, but appears to have been observed at 1329 cm−1 [8], the transitions becoming allowed through Coriolis perturbation. v5 has apparently been observed in the Raman also at 1333 cm−1 [9].
No account was taken of the effect of centrifugal distortion in this analysis. Since only levels with low J values were used in the analysis, the effect of this correction on the rotational constants was minimized. No systematic differences between the observed and calculated spectra were noticed until rather high J values were reached. In these regions of the absorption serious overlapping of transitions make the unique assignment of transitions to observed absorption peaks doubtful.
Although no statistical analysis of the data was made, the excellent agreement between the observed and calculated ΔF2 values for the ground state, and the sensitivity of the calculated ΔF2 values to values of the rotational constants seem to indicate a probable error of the order of ±0.002 cm−1 for each of the ground state constants. The agreement between the calculated and observed values of ΔF2 for the ground state may be seen in table 1.
Table 1.
Ground state
| A-type band
|
C-type band
|
||||
|---|---|---|---|---|---|
| Calculated | Observed | Calculated | Observed | ||
|
|
|
|
|
|
|
| cm−1 | cm−1 | cm−1 | cm−1 | ||
| 000–202 | 19.513 | 19.55 | 000–220 | 23.911 | 23.92 |
| 101–303 | 32.103 | 32.10 | 101–321 | 37.524 | 37.53 |
| 111–313 | 31.547 | 31.55 | 111–331 | 41.285 | 41.29 |
| 110–312 | 33.791 | 33.79 | 110–330 | 40.862 | 40.88 |
| 202–404 | 44.373 | 44.38 | 202–422 | 51.644 | 51.64 |
| 212–414 | 43.976 | 43.98 | 212–432 | 55.068 | 55.06 |
| 211–413 | 46.958 | 46.96 | 211–431 | 53.908 | 53.91 |
| 221–423 | 45.754 | 45.76 | 221–441 | 58.305 | 58.27 |
| 220–422 | 47.277 | 47.25 | 220–440 | 58.165 | 58.16 |
| 303–505 | 56.501 | 56.50 | 303–523 | 66.378 | 66.37 |
| 313–515 | 56.300 | 56.28 | 313–533 | 69.201 | 69.20 |
| 312–514 | 59.715 | 59.69 | 312–532 | 67.212 | 69.23 |
| 322–524 | 58.539 | 58.54 | 322–542 | 71.733 | 71.73 |
| 321–523 | 60.967 | 60.96 | 321–541 | 71.127 | 71.12 |
| 331–533 | 59.463 | 59.47 | 331–551 | 75.463 | 75.43 |
| 330–532 | 60.141 | 60.13 | 330–550 | 75.431 | 75.43 |
| 404–606 | 68.637 | 68.61 | 404–624 | 81.651 | 81.64 |
| 414–616 | 68.559 | 68.53 | 414–634 | 83.676 | 83.63 |
| 413–615 | 72.042 | 71.98 | 413–633 | 81.042 | 81.03 |
| 423–625 | 71.147 | 71.11 | 423–643 | 85.399 | 85.38 |
| 422–624 | 74.350 | 74.33 | 422–642 | 83.950 | 83.93 |
| 432–634 | 72.585 | 72.55 | 432–652 | 88.728 | 88.65 |
| 431–633 | 74.092 | 74.07 | 431–651 | 88.532 | 88.52 |
| 441–643 | 72.848 | 72.81 | 441–661 | 92.665 | 92.56 |
| 440–642 | 73.062 | 73.02 | 440–660 | 92.660 | 92.56 |
The constants for the excited states of these bands cannot be determined as precisely as those for the ground state with the available ΔF2 values, but ±0.005 cm−1 would seem to be a generous estimate of the probable error in the constants for these states. The calculated and observed ΔF2 values for the excited states are compared in tables 2 and 3.
Table 2. C-type band .
A=4.255 cm−1 B=3.590 cm−1 C=3.151 cm−1
| ΔF′2 | Calc | Obs |
|---|---|---|
|
|
|
|
| cm−1 | cm−1 | |
| 000–220 | 23.918 | 23.96 |
| 101–321 | 37.927 | 37.94 |
| 111–331 | 41.086 | 41.12 |
| 110–330 | 40.685 | 40.70 |
| 202–422 | 52.436 | 52.44 |
| 212–432 | 55.238 | 55.25 |
| 211–431 | 54.163 | 54.18 |
| 221–441 | 57.925 | 57.91 |
| 220–440 | 57.778 | 57.80 |
| 303–523 | 67.521 | 67.51 |
| 313–533 | 69.726 | 69.74 |
| 312–532 | 67.944 | 67.93 |
| 322–542 | 71.753 | 71.74 |
| 321–541 | 71.131 | 71.15 |
| 331–551 | 74.891 | 74.82 |
| 330–550 | 74.854 | 74.82 |
| 404–624 | 83.064 | 82.95 |
| 414–634 | 84.534 | 84.48 |
| 413–633 | 82.268 | 82.15 |
| 423–643 | 85.784 | 85.82 |
| 422–642 | 84.431 | 84.42 |
| 432–652 | 88.552 | 88.51 |
| 431–651 | 88.324 | 88.32 |
| 441–061 | 91.899 | 91.83 |
| 440–660 | 91.890 | 91.83 |
Table 3. A-type band .
A=4.254 cm−1 B=3.654 cm−1 C=3.019 cm−1
| Calc | Obs | |
|---|---|---|
|
|
|
|
| cm−1 | cm−1 | |
| 000–202 | 19.715 | 19.70 |
| 101–303 | 33.132 | 32.11 |
| 111–313 | 31.617 | 31.62 |
| 110–312 | 34.688 | 34.68 |
| 202–404 | 44.163 | 44.17 |
| 212–414 | 43.916 | 43.91 |
| 211–413a | 47.748 | 47.51 |
| 221–423 | 46.516 | 46.54 |
| 220–422 | 48.979 | 48.96 |
| 303–505a | 56.159 | 56.51 |
| 313–515 | 56.080 | 56.07 |
| 312–514a | 60.078 | 59.35 |
| 322–524 | 59.144 | 59.11 |
| 321–523a | 62.809 | 62.70 |
| 331–533 | 60.862 | 60.86 |
| 330–532 | 62.271 | 62.59 |
Perturbed levels.
The constants determined for the three vibrational levels are given in table 4. The band origins were determined from the best fit between the observed and calculated spectra for low J values.
Table 4.
Rotational and vibrational constants
| Ground state | v3+v9 | v3+v6 | |
|---|---|---|---|
|
|
|
|
|
| cm−1 | cm−1 | cm−1 | |
| A | 4.303 | 4.254 | 4.255 |
| B | 3.504 | 3.654 | 3.590 |
| C | 3.049 | 3.019 | 3.151 |
| v0 | 2671.67 | 4425.61 |
Figure 1. The A-type hand of CH2D2 at 2672 cm−1.
The identification given is the ground state designation of for R0,1 transitions to the high wave number side of the band origin, and for P0, transitions to the low wave number side of the band origin.
Figure 2. The C-type band of CH2D2 at 4426 cm−1.
The identification given is the ground state designation of for R1,0 transitions to the high wave number side of the band origin, and for , 0 transitions to the low wave number side of the band origin.
Footnotes
This work was supported in part by the Division of Chemistry, Atomic Energy Commission.
Figures in brackets indicate the literature references at the end of this paper.
5. References
- 1.Among the more recent high resolution analyses are those of:; Hecht KT. J Mol Spectroscopy. 1960;5:355, 390. [Google Scholar]; Thomas MA, Welsh HL. Can J Phys. 1960;38:1291. [Google Scholar]; Moret-Bailly J. Cahiers de Phys. 1961;130–131:1. [Google Scholar]; Olafson RA, Thomas MA, Welsh HL. Can J Phys. 1961;39:419. [Google Scholar]; Allen HC, Jr, Plyler EK. J Research NBS. 1959;63:145. doi: 10.6028/jres.063A.007. [DOI] [PMC free article] [PubMed] [Google Scholar]; Richardson EH, Brodersen S, Krause L, Welsh HL. J Mol Spectroscopy. 1962;8:406. [Google Scholar]
- 2.Allen HC., Jr Phil Trans Roy Soc London. 1961;A253:335. [Google Scholar]
- 3.Gailar N, Plyler EK. J Research NBS. 1955;45:102. [Google Scholar]
- 4.E. K. Plyler, L. R. Blaine, and E. D. Tidwell, ibid., 55, 279 (1955).
- 5.Townes CH, Schawlow AL. Microwave Spectroscopy. McGraw-Hill; New York: 1955. [Google Scholar]
- 6.Bragg JK, Golden S. Phys Rev. 1949;75:735. [Google Scholar]
- 7.Cross PC, Hainer RM, King GW. J Chem Phys. 1944;12:210. [Google Scholar]
- 8.Wilmshurst JK, Bernstein HJ. Can J Chem. 1957;35:226. [Google Scholar]
- 9.Mac Wood GE, Urey HC. J Chem Phys. 1936;4:402. [Google Scholar]


