Free Access
Issue
A&A
Volume 564, April 2014
Article Number A82
Number of page(s) 5
Section Atomic, molecular, and nuclear data
DOI https://doi.org/10.1051/0004-6361/201322800
Published online 10 April 2014

© ESO, 2014

1. Introduction

Almost 30 nitrile molecules, which contain the cyano group (CN), have been detected in the interstellar medium (ISM) so far. The simplest of these nitriles, the cyano radical CN, was found in 1940 (McKellar 1940; Hakobian & Crutcher 2011). The most complex of all nitriles, HC11N, having a total of 13 atoms, was detected in 1997 (Bell et al. 1997). The vast majority of these nitriles have been observed in the dense molecular clouds Sgr A and B, and TMC-1 by means of their rotational spectra. The formation mechanisms of many of these compounds are not well understood.

Two interesting nitrile series of some relevance to the present study have been detected in the ISM. The first of these has the general formula H(CC)n-CN, where n = 0 − 5 (Snyder & Buhl 1971; Turner 1971; Avery et al. 1976; Broten et al. 1976; Kroto et al. 1978; Little et al. 1978; Winnewisser & Walmsley 1978; Broten et al. 1978; Bell et al. 1997). The hydrogen atom is replaced by a methyl group in the second series CH3(CC)n-CN, where n = 0 − 2 (Solomon et al. 1971; Broten et al. 1984; Snyder et al. 2006).

Recently, cyanoallene CH2=C=CH-CN was observed in the dense molecular cloud TMC-1 (Lovas et al. 2006). This compound seems to be more abundant than its methyl cyanoacetylene isomer, CH3CC-CN. The present work deals with the corresponding methyl substituted cyanoallene, namely, 4-methylcyanoallene or 2, 3-pentadienylnitrile CH3CH=C=CH-CN. In this molecule, one hydrogen atom in cyanoallene has been replaced by a methyl group. We suggest that 4-methylcyanoallene is an interesting candidate for a future radioastronomical detection. There are several reasons for this suggestion. First, CH3CH=C=CH-CN is chemically closely related to the interstellar molecule CH2=C=CH-CN. Second, the existence in the ISM of the H(CC)n-CN and CH3(CC)n-CN series referred to above indicates that replacement of a hydrogen atom attached to an unsaturated carbon atom with a methyl group may occur not only for the cyanoacetylene series, but perhaps for cyanoallenes as well.

Moreover, a very interesting property of 4-methylcyanoallene is its chirality, the fact that this compound exists in two non-superimposable mirror-image forms (see Fig. 1). Most of the amino acids and sugars that are considered as the building blocks of life on the Earth are chiral and only one enantiomer is present in the known living systems. To date, the only detected extraterrestrial chiral compounds came from meteorites after an aqueous acidic treatment to extract the organic compounds. Therefore, some amino acids with an enantiomeric excess have been found (Pizzarello & Groy 2011). In the ISM neither branched nor chiral compounds have been detected. Even if it is not possible to distinguish between different enantiomers in astronomical radio spectra, the detection of chiral compounds in the ISM will be helpful to understand the origin of the chiral species encountered in meteorites and to demonstrate that more complex compounds than those observed to date are present in the ISM. Many microwave spectra of chiral compounds have been assigned. Searches for two of them, propylene oxide (Cunningham et al. 2007) and aminopropionitrile (Møllendal et al. 2012) were unsuccessful.

thumbnail Fig. 1

Models of the two mirror-image (enantiomers) forms of 4-methylcyanoallene where the blue atom is a nitrogen atom. The two forms have identical rotational spectra and spectroscopic constants.

Open with DEXTER

The abundance of CH3CH=C=CH-CN, a compound consisting of 11 atoms, is not expected to be high in the ISM. Fortunately, its dipole moment is quite large (~4.5 D; see below) resulting in relatively intense rotational transitions, which should facilitate its detection using the spectroscopic material presented in this article. The increased sensitive and spectral resolution of radiotelescopes, for example ALMA or the Jansky VLA, may lead to the identification of this molecule in the ISM.

2. Experimental

We purchased from the Aldrich Company the 3-Bromo-1-butyne used in the synthesis without further purification. The synthesis of 4-methylcyanoallene was performed as described in the literature (Brandsma & Verkruijsse 1981) and obtained in 83% yield. The synthesis and nuclear magnetic resonance (NMR) data are given in the Appendix.

The Stark-modulated spectrometer of the University of Oslo that was used in this study has been described elsewhere (Møllendal et al. 2005, 2006). This spectrometer has a resolution of 0.5 MHz and measures the frequency of isolated transition with an accuracy of 0.1 MHz. The spectrometer employs a 50 kHz home-built Stark generator. The microwave radiation is generated using a 1730B Systron Donner frequency synthesizer operating in the 226.5 GHz frequency range. Several Millitech frequency multipliers are used to generate radiation in the 26.5116 GHz spectral interval. The lock-in amplifier is a Perkin Elmer model 5209. Double resonance radio-frequency microwave experiments (RFMWDR), similar to those performed by Wodarczyk and Wilson (Wodarczyk & Wilson 1971), have been employed to unambiguously assign several transitions using the equipment described elsewhere (Leonov et al. 2000). The radio frequency source is a Rohde & Schwarz SML01 signal generator operating in the 9 kHz1.1 GHz spectral region. An EIN Model 503L amplifier provides 3 W linear amplification of the radio signals between 2 and 510 MHz. Mixing of the radio signal with the Stark modulation signal is provided using a Hewlett-Packard 10 514 mixer. The pressure in the spectrometer cell was approximately 10 Pa during measurements. The temperature was maintained at − 30 °C by cooling the 2 m Hewlett-Packard absorption cell with small portions of dry ice when recording the spectrum in order to enhance spectral intensities.

3. Microwave spectrum and assignment

Quantum-chemical calculations were performed to obtain estimates of useful spectroscopic parameters employing the Gaussian09 program package (Frisch et al. 2009) running on the Abel Linux cluster at the University of Oslo. Becke’s three-parameter hybrid functional (Becke 1988) employing the Lee, Yang, and Parr correlation functional (B3LYP; Lee et al.) was used in the density functional theory calculations (DFT) to calculate rotational constants and Watson’s S-reduction quartic and sextic centrifugal distortion constants (Watson 1977) observing the precautions of McKean et al (McKean et al. 2008). Very high level ab initio coupled-cluster calculations with singlet and doublet excitations (CCSD; Purvis & Bartlett, Scuseria et al. 1988) were also undertaken to calculate the structure, rotational constants, and dipole moment components along the principal inertial axes. Peterson and Dunning’s correlation consistent cc-pVTZ basis set was used in the calculations (Peterson & Dunning 2002).

The CCSD rotational constants of CH3CH=C=CH-CN (Table 1) show that this compound is close to the prolate rotor limit, with Ray’s asymmetry parameter κ ≈ − 0.986 (Ray 1932). The major dipole moment component is found along the a-axis (~4.36 D), whereas μb ~ 1.57 and μc ~ 0.39 D.

Table 1

Spectroscopic constants1 of CH3CH=C=CH-CN.

The survey spectrum revealed the expected pattern of strong pile-ups of a-type R-branch transitions separated by almost exactly the sum of the B and C rotational constants. These pile-ups generally have a complicated and crowded fine structure due to the presence of a number of vibrationally excited states, and to a more or less resolved K-1 fine structure. Overlapping transitions therefore occur frequently. A portion of the J = 35 ← 34 transitions illustrating this complexity is shown in Fig. 2. The most intense lines here belong to the ground vibrational state, while the smaller lines belong to vibrationally excited states. Most of the prominent vibrationally excited states are found at higher frequencies (not shown in this figure).

thumbnail Fig. 2

A portion of the J = 35 34 ground vibrational state aR-transitions of the rotational spectrum of CH3CH=C=CH-CN taken at a Stark field of 50 V/cm. The intensity (y-axis) is in arbitrary units. The transitions have positive amplitudes, whereas the Stark lobes have negative amplitudes. The numbers above the assigned peaks refer to the K-1 pseudo quantum number. The unlabeled and most intense line to the extreme left consists of the unresolved K-1 = 6 and 8 transitions. The line corresponding to K-1 = 17 is overlapped by an excited state transition as well as by Stark lobes. The minor lines presumably belong to vibrationally excited states. Stark lobes interfere with the other lines to a very minor degree, but are not completely absent.

Open with DEXTER

The aR-transitions of the ground state were readily assigned. Several of these assignments were confirmed by RFMWDR experiments. The Stark-modulation properties of the transitions were also useful in this respect. The b-type lines are significantly weaker than the a-type transitions and they were assigned by a trial and error procedure. No c-type transitions were unambiguously identified, presumably because they are too weak, which is consistent with a small μc (see above). No resolved hyperfine structures caused by 14N nuclear quadrupole interaction or tunneling of the methyl group were seen.

Table 2

Predictions of spectral properties of the strongest transitions of CH3CH=C=CH-CN above 40 GHz at 10 K.

Table 3

Rotational spectrum of CH3CH=C=CH-CN in the ground vibrational state.

Table 4

Rotational spectrum of CH3CH=C=CH-CN in the lowest vibrationally excited state.

Sørensen’s program ROTFIT (Sørensen 1967) was used for a weighted least-squares fit of the assigned lines employing Watson’s S-reduction and Ir-representation Hamiltonian (Watson 1977). Sharp and well-isolated aR-transitions with normal Lorentzian shapes and not significantly perturbed by Stark lobes of neighboring lines or overlapped by vibrationally excited-state transitions were assigned an uncertainty of ± 0.10 MHz. Weaker lines in the same category were included in the fit, but larger uncertainties were assigned to them. The b-type transitions were generally weaker than the a-type lines and weights of at least ± 0.15 MHz were used for them. Because overlapping of transitions occurs frequently, only well-isolated lines were included in the fit. Ultimately, a total of 654 aR, bQ, and bR-branch transitions with values of J between 5 and 52 and K-1 between 0 and 29 of the ground vibrational state were used in the fit. These transitions together with the frequencies of the upper states are listed in Table 3.

The inverse squares of the uncertainties were used as weights in the least-squares fit. It was possible to obtain significant values for the five quartic and three of the seven sextic Watson constants, HJ, HJK, and HKJ, as shown in Table 1. The values of the remaining sextic constants, HK, h1, h2, and h3, were preset at zero in the least-squares fit.

It can be seen from Table 1 that the experimental and CCSD rotational constants agree to better than 0.7%, which is very satisfactory given the different definitions of the two sets of constants. The experimental constants are effective constants, while the CCSD constants have been calculated from an approximate equilibrium structure. The B3LYP quartic centrifugal distortion constants deviate by as much as 24% (dK), while no comparison is warranted for the sextic constants because of their inaccuracy.

The spectrum of the first excited state of the lowest bending vibration was assigned in the same manner as described for the ground vibrational state (see Table 4). The spectroscopic constants obtained from 342 transitions are included in Table 1. The HJK, and HKJ centrifugal distortion constants were kept at their ground-state values in the least squares fit in this case. Relative intensity measurements yielded 107(25) cm-1 for this vibration, compared to 97 cm-1 (not given in Table 1) obtained for the anharmonic frequency in the B3LYP calculations. The rotational energies of the upper states are not given in this table, because they are very small compared to the vibrational energy (107(25) cm-1 above the ground-state energy). The intensities of rotational transitions of this excited vibrational state are therefore very dependent on this vibrational frequency.

The spectroscopic parameters obtained in the present work should be able to predict very accurately the rest frequencies with associated intensities that can be used for a potential identification of CH3CH=C=CHCN, as the following example will demonstrate. The cold TMC-1 cloud with a temperature of roughly 10 K should be a potential source for this compound. The ALMA observatory will ultimately operate in the 40959 GHz spectral region (ALMA Sciences 2013; Di Francesco et al. 2013) and we have therefore calculated the rest frequencies of the 54 transitions that have the largest integrated intensities (Townes & Schawlow 1955) in this spectral interval at 10 K. All these transitions, which are listed in Table 2, are a-type R-branch transitions with the quantum number J between 12 and 22, and with K-1 equal to or less than 3. They are predicted to have uncertainties (one standard deviation) no larger than ±0.014 MHz. It can be seen from this table that all the strongest transitions occur below 67 GHz, which is due to the low temperature which produces unfavorable Boltzmann factors for energy levels with even higher J and K-1.

4. Conclusion

The rotational spectra of the ground and of the first excited vibrational state of the lowest bending vibration of the chiral compound CH3CH=C=CH-CN have been investigated in the 13166 GHz spectral region. We have assigned 654 transitions of the ground state and 379 transitions of the first excited state of the lowest bending vibration. The dipole moment has been calculated at the high CCSD/pVTZ level of theory. The spectroscopic constants obtained from the least-squares fit of these transitions should be able to predict very accurately the rest frequency of transitions not appearing in the investigated spectral range. The accurate rest frequencies of the 54 strongest transitions at 10 K and which are within the range of the ALMA Band 1 project and/or with the Jansky VLA (Di Francesco et al. 2013), are listed in Table 2. It is hoped that the present spectral analysis of CH3CH=C=CH-CN will facilitate its detection in the ISM, provided it is present in a suitable concentration.

Acknowledgments

We are grateful to Anne Horn for her skillful assistance. S. C., J.-C. G. and Y.T. acknowledge the “Programme de Physique et Chimie du Milieu Interstellaire” and the “Programme National de Planétologie” (PCMI and PNP INSU-CNRS). In addition, J.-C.G. and Y.T. acknowledge the “Centre National d’Études Spatiales” (CNES) for financial support. This work has been supported by the Research Council of Norway through a Centre of Excellence Grant (Grant No. 179568/V30). It has also received support from the Norwegian Supercomputing Program (NOTUR) through a grant of computer time (Grant No. NN4654K).

References

Appendix A: Details of the synthesis and NMR spectrum of 4-methylcyanoallene (CH3CH=C=CH-CN)

Ethanol (5 mL), water (0.3 mL), finely powdered cuprous cyanide (0.30 g, 3.35 mmol) and 3-bromo-1-butyne (1.33 g, 10 mmol) were placed in a two-necked flask, equipped with a nitrogen inlet and a stirring bar. The mixture was warmed to 55 °C and a solution of potassium cyanide (0.65 g, 10 mmol) in water (2 mL) was added drop-wise. The temperature of the mixture was maintained close to 60 °C throughout the period of addition and then the mixture was heated to 70 °C under vigorous stirring for 30 min. After cooling to 20 °C, 10 mL of ice-water was added and the organic product

was extracted seven times with diethyl ether (7 × 10 mL). The organic phase was washed twice with saturated ammonium chloride solution, dried over magnesium sulfate and most of the diethyl ether was removed by distillation at atmospheric pressure. The 4-methylcyanoallene was purified on a vacuum line (0.1 mbar) by selective condensation in a trap immersed in a bath cooled at −50 °C. Yield: 0.66 g (8.34 mmol), 83%. 1H NMR (CDCl3, 400 MHz) δ 1.68 (dd, 3H, 3JHH = 7.5 Hz, 5JHH = 3.3 Hz, CH3); 5.11 (dq, 4JHH = 6.6 Hz, 5JHH = 3.3 Hz, CHCN); 5.61 (qd, 3JHH = 7.5 Hz, 4JHH = 6.6 Hz, CHMe). 13C NMR (CDCl3, 100 MHz) δ 12.3 (1JCH = 131.0 Hz (q), CH3); 66.4 (1JCH = 185.8 Hz (d), HC-CN); 91.7 (1JCH = 165.1 Hz (d), C-Me); 113.5 (s, CN); 215.7 (s, C=C=C). The compound 4-methylcyanoallene is a colorless liquid with a vapor pressure about 100 Pa at room temperature.

All Tables

Table 1

Spectroscopic constants1 of CH3CH=C=CH-CN.

Table 2

Predictions of spectral properties of the strongest transitions of CH3CH=C=CH-CN above 40 GHz at 10 K.

Table 3

Rotational spectrum of CH3CH=C=CH-CN in the ground vibrational state.

Table 4

Rotational spectrum of CH3CH=C=CH-CN in the lowest vibrationally excited state.

All Figures

thumbnail Fig. 1

Models of the two mirror-image (enantiomers) forms of 4-methylcyanoallene where the blue atom is a nitrogen atom. The two forms have identical rotational spectra and spectroscopic constants.

Open with DEXTER
In the text
thumbnail Fig. 2

A portion of the J = 35 34 ground vibrational state aR-transitions of the rotational spectrum of CH3CH=C=CH-CN taken at a Stark field of 50 V/cm. The intensity (y-axis) is in arbitrary units. The transitions have positive amplitudes, whereas the Stark lobes have negative amplitudes. The numbers above the assigned peaks refer to the K-1 pseudo quantum number. The unlabeled and most intense line to the extreme left consists of the unresolved K-1 = 6 and 8 transitions. The line corresponding to K-1 = 17 is overlapped by an excited state transition as well as by Stark lobes. The minor lines presumably belong to vibrationally excited states. Stark lobes interfere with the other lines to a very minor degree, but are not completely absent.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.