EDP Sciences
Free Access
Issue
A&A
Volume 577, May 2015
Article Number A91
Number of page(s) 6
Section Atomic, molecular, and nuclear data
DOI https://doi.org/10.1051/0004-6361/201525790
Published online 08 May 2015

© ESO, 2015

1. Introduction

The recent detection of increasingly complex molecules such as the first detection of ethyl formate (CH3CH2OCOH) toward Sgr B2(N) (Belloche et al. 2009, 2013) as well as the discovery of methyl acetate (CH3COOCH3) and the gauche conformer of ethyl formate (g-CH3CH2OCOH) in Orion KL (Tercero et al. 2013) makes the acetate containing species potential molecules in the interstellar medium. Among them, vinyl acetate (CH3C=OOCH=CH2) is a logical molecule to search for since it is related to methyl acetate and it is the simplest unsaturated carboxylic ester after vinyl formate. The line survey of Orion KL from 80 to 280 GHz (Tercero et al. 2010) using the IRAM 30 m telescope presented initially more than 8000 unidentified lines. Nearly 4000 of them have been identified as lines arising from isotopologues and vibrationally excited states of abundant species (Tercero et al. 2011, 2012; Daly et al. 2013; Demyk et al. 2007; Margulès et al. 2009; Carvajal et al. 2009; López et al. 2014) or other organic molecules (Tercero et al. 2013; Kolesniková et al. 2013, 2014). The study of a cloud such as Orion KL is a good template that could provide important information on the formation of complex organic molecules on the grain surfaces and/or in the gas phase. The ALMA observations are also bringing very sensitive data from the interstellar medium. Therefore, it is important to gather spectroscopic data on potentially detectable organic molecules. The work done by Tercero and collaborators is a first and mandatory step towards interpreting the complex spectra that ALMA will provide for hot cores such as Orion. Taking into account the large number of lines to be identified in Orion, the collaboration between astronomers and spectroscopists is the only way to obtain the maximum scientific output from the line surveys that will be obtained with ALMA.

It was only recently that the rotational spectrum and effective molecular structure of vinyl acetate was first studied. Velino et al. (2009) measured its spectrum between 61 and 77 GHz using the free-jet millimeter absorption technique. The quantum chemical calculations performed by those authors showed that the so-called cis-trans (sp,ap in IUPAC notation) conformer, presented in Fig. 1, is the most stable one. Recently, another work on vinyl acetate in the microwave range from 6 to 40 GHz was performed (Nguyen et al. 2014).

Although the above results provided valuable information about the low-frequency rotational spectrum of vinyl acetate, they cannot be used to accurately predict its frequencies in the millimeter wave domain. There is always some degree of uncertainty involved in predictions of the positions of higher frequency transitions, especially for molecules exhibiting a large amplitude motion with a rather small torsional barrier like vinyl acetate. Possible interstellar detection of vinyl acetate in the millimeter line survey of Orion or another interstellar source should thus be based on transitions measured directly in the laboratory or transitions predicted from a data set that includes higher frequency transitions. Therefore, the objective of the present work is to gather an atlas of line positions and intensities for vinyl acetate up to 305 GHz extending the range of J and Ka to 75 and 16, respectively, and providing the precise set of the spectroscopic constants that could be used to search for this species in the interstellar medium.

thumbnail Fig. 1

The lowest energy conformer of vinyl acetate (sp, ap) in the principal axis system.

Open with DEXTER

2. Experiments

Vinyl acetate (b.p.: 72°C) was purchased from Sigma-Aldrich and used without any further purification. To record the rotational spectrum in the 618 GHz frequency range, the sample was loaded into the reservoir of two pulsed nozzles and probed by broadband CP-FTMW spectroscopy. Details about the CP-FTMW spectrometer can be found elsewhere (Mata et al. 2012). Neon carrier gas at a pressure of 2 bar was used for the pulsed jet expansion. To obtain the rotational spectrum of vinyl acetate, 230 000 individual free induction decays (four FIDs on each valve cycle) were acquired. A fast Fourier transform with a Kaiser-Bessel window was used to convert the averaged time domain spectrum to the broadband frequency domain spectrum shown in Fig. 2a.

thumbnail Fig. 2

a) Broadband chirped pulse Fourier transform microwave spectrum (CP-FTMW) from 6 to 18 GHz. b) A section of the millimeter wave spectrum of vinyl acetate showing the assignments of the A- and E- components for the ground torsional state transitions.

Open with DEXTER

thumbnail Fig. 3

Loomis-Wood-type plot obtained from the AABS programs package (Kisiel et al. 2005) showing the identification of the stripes corresponding to the sequences of the b-type R-branch transitions for , and 4. Each sequence has two components, A and E, due to the methyl internal rotation. Rotational transitions are lined up to the central frequencies νcent of the A-component transitions with the rotational quantum numbers indicated on the right side of the diagram. The frequency distance Δν from the central frequency is documented on the x-axis. Each A-E doublet going to the left from the leading one at Δν = 0 corresponds to increasing Ka and decreasing J transitions. The highlighted horizontal stripe represents the part of the spectrum documented in Fig. 2b. For J′′< 60, the degeneracies for the energy levels with Ka = 3 and Ka = 4 are lifted and two separate sequences of (the left one) and (the right one) transitions are recognizable for both A- and E-components.

Open with DEXTER

thumbnail Fig. 4

Lines of vinyl acetate, CH3COOCHCH2, towards the compact ridge of Orion KL. The black histogram is the averaged spectrum over 5 × 5 pixels around the methyl acetate peak (see text) of the ALMA SV data (see text for references); these data have a 0.5 MHz of spectral resolution. The thin red line is the synthetic spectrum assuming a column density of 2 × 1015 cm-2 for vinyl acetate (A+E species). The thin blue line is the model for vinyl acetate with a column density of 1 × 1015 cm-2 (A+E species); this model corresponds to the given column density for the tentative detection of vinyl acetate (see text). A vLSR of 7 km s-1 is assumed.

Open with DEXTER

Rotational spectra in the millimeter wave region (125305 GHz) were recorded by a recently constructed millimeter wave spectrometer at the University of Valladolid. All details concerning the experimental setup (synthesizer, cascade frequency multipliers, detectors) can be found in Daly et al. (2014). Room temperature measurements were carried out at a pressure of 2030 μbar. All spectra were recorded in 1GHz sections using the frequency modulation technique with 2f lock-in detection, where f is the modulation frequency of 10.2 kHz, and modulation depth between 3050 kHz. Frequency accuracy is estimated to be better than 50 kHz.

3. Spectroscopic analysis

The rotational spectrum of vinyl acetate has two characteristic features. First, the calculated non-zero dipole moment components of μa = −0.05 D, μb = −1.62 D, and μc = 0.06 D from Velino et al. (2009) points out that only b-type transitions are relevant. We should note that the ab initio value of the out-of-plane component of the dipole moment μc is likely to be an calculation artefact. Second, vinyl acetate, like other complex organic molecules, exhibits a large amplitude motion, i.e., the torsion of the methyl group relative to the rest of the molecule. The threefold barrier hindering this motion is rather small (V3 = 152.117 (39) cm-1) leading to relatively large splittings of the rotational lines into A- and E-components even in the ground torsional state vt = 0. While only a limited number of A-E doublets of the b-type R-branch transitions could be measured in the microwave 618 GHz region (see Fig. 2a), these transitions dominate in the millimeter wave region. A section of the millimeter wave spectra measured is shown in Fig. 2b, where the A-E doublets for lowest values of the Ka quantum numbers are assigned. The leading A and E lines consist of a pair of b-type R-branch transitions between degenerate Ka = 0 and Ka = 1 levels, see Fig. 2b. Successive A-E doublets going from the leading one to the left are for increasing Ka and decreasing J transitions and reveal the same type of degeneracies. Figure 2b also illustrates the dependence of the A-E splitting magnitude on the Ka quantum number.

Assignment of the low J transitions from 6 to 18 GHz was straightforward, since the supersonic expansion technique allows us to cool down the sample and to observe only the rotational transitions from the lowest energy levels (see Fig. 2a). Transitions measured in the CP-FTMW spectra (0 ≤ J ≤ 12 and Ka ≤ 4) and those from Velino et al. (2009) (3 ≤ J ≤ 25 and Ka ≤ 5) were weighted with an experimental uncertainty of 5 kHz and 40 kHz, respectively, and analyzed together using the internal rotation program BELGI-Cs (Hougen et al. 1994; Kleiner 2010). The principal advantage of the RAM general approach used in the BELGI-Cs code is that takes into account simultaneously A- and E-symmetry states and makes it possible to reproduce the experimental data within the experimental accuracy from the microwave to the submillimeter wave range. This first fit enabled us to provide new predictions to facilitate the extension of the assignments of the vinyl acetate rotational transitions into the millimeter wave region. The millimeter wave spectrum of vinyl acetate exhibits higher line density than CP-FTMW spectrum not only because of the higher J and Ka transitions from the vt = 0 state, but also from the low-lying excited vibrational states sufficiently populated at the room temperature of the experiment. Loomis-Wood-type plots in the AABS package (Kisiel et al. 2005) were used to guide a search of both A and E lines. An example of this plot is shown in Fig. 3 where all transitions are lined up to the central frequencies of the A-component transitions and each sequence for both A- and E-components then forms a visible stripe running from the lower to the upper values of J quantum numbers.

For consistency, apart from the 62 CP-FTMW and 2220 millimeter wave transitions measured in this work, 72 additional transitions from Velino et al. (2009) and 154 transitions measured by Nguyen et al. (2014) were included in the final fit. Hence, 1378 transitions for the A-component and 1130 transitions for the E-component were fit with an overall (unitless) standard deviation of 0.905 (root mean square deviation of 5.1 kHz for the FTMW lines and 42.2 kHz for the millimeter wave lines) using 24 molecular parameters. Table 1 presents all 24 floated torsion-rotation parameters up to the 6th order of RAM model. Among the floated parameters, there are three rotational constants ARAM, BRAM, and CRAM as well as the Dab parameter multiplying the JaJb + JbJa operator. The Dab parameter comes from the use of the rho-axis system which is not the principal axis system. Dab is directly related to the angle θRAM between the rho-axis system and the principal axis system (Kleiner 2010). For vinyl acetate, θRAM = 11.83°. Table 1 also provides the values for the centrifugal distortion parameters in the rho-axis system, the barrier height V3, the value of the internal rotation constant F and the coupling constant between internal and global rotation ρ. Because of the rather low value of the barrier and the extended dataset involving higher J values, it was possible to simultaneously float both V3 and F. Finally, 11 higher parameters describing the interaction between internal and global rotation (Dac, Δab, k2, k5, Fv, Gv, c1, c2, c4, dab, DabJ) had to be introduced to achieve the fit reproducing the data within the experimental uncertainty. Line assignments, observed frequencies νobs, calculated frequencies νcalc with the corresponding uncertainties, νobsνcalc values, and references of the data sources included in the final fit for the torsional ground state of vinyl acetate are presented in Table 2.

Table 1

Molecular constants in the rho-axis system of vinyl acetate obtained from the global fit using program BELGI-Cs.

4. Calculation of intensities

The line strength calculation is reported here because it is an absolute prerequisite for correct molecular identification in the interstellar medium. For the line strength calculation, the procedure reported for the RAM formalism has already been explained in detail (Hougen et al. 1994; Kleiner 2010). The only values available for the dipole moment components in the principal axis system of vinyl acetate are those obtained from ab initio calculations (see Table 1 in Velino et al. 2009). Under the assumption that the line strengths are only driven by the permanent electric dipole components and have no torsional dependencies (Hougen et al. 1994), the components of the dipole moment in the RAM axis system of μa(RAM) = −0.38 D, μb(RAM) = −1.5755 D were used. These values were obtained by rotating the dipole moment under the angle θRAM = 11.83° between RAM and PAM axis systems. A very small value of the out-of-plane dipole moment component, μc, was neglected. Calculated line strengths together with lower and upper state energies of vinyl acetate in vt = 0 are presented in Table 2.

5. Radioastronomical observations

5.1. IRAM 30 m data

Three species containing the vinyl group have been identified in the interstellar medium: vinyl cyanide (Gardner & Winnewisser 1975), vinyl alcohol (Turner & Apponi 2001), and propenal (Hollis et al. 2004). Although the last two have not been detected towards Orion KL, the first, vinyl cyanide, presents a large abundance in this source (López et al. 2014). Therefore, after the detection of methyl acetate in Orion KL (Tercero et al. 2013) and taking into account the large abundance of vinyl species in this source (López et al. 2014), vinyl acetate seems to be a potential candidate for detection in the millimeter line survey with the IRAM 30 m telescope towards Orion KL (Tercero et al. 2010) ranged from 80 to 307 GHz (new data from 280 to 307 GHz have recently been observed). In order to proceed with the search for vinyl acetate, the new spectroscopic data from this work were implemented into the MADEX code (Cernicharo 2012). However, we did not find this species above the confusion limit in those observations. In order to derive an upper limit to its column density, we assumed the same physical conditions as those provided in Tercero et al. (2013) for methyl acetate. Therefore, we assume that most of the vinyl acetate emission should come from the so-called compact ridge component of this source (see, e.g., Blake et al. 1987, for the different spectral components found in Orion KL with single-dish observations). Using the MADEX code we found a total upper limit to its column density (A+E species) of (4 ± 2) × 1014 cm-2 assuming LTE approximation, vLSR = 8 km s-1, Δv = 3 km s-1, and the following physical properties for the compact ridge: TK = 150 K, a source diameter of 15, and a location placed 7 with respect to the pointing position of the 30 m data (IRc2). Hence, using the column density value of methyl acetate derived by Tercero et al. (2013), we found an abundance ratio between both species of N(CH3COOCH3)/N(CH3COOCHCH2) ≥ 11. To compare with other results of the ratio between methyl/vinyl species, the derived value is 1.3 times higher than the total abundance ratio between methyl and vinyl cyanide given in López et al. (2014) in the same source but in another well known component, the so-called hot core. Other interesting ratios that could give us information about the relative abundance of vinyl acetate are those between the methyl/ethyl and ethyl/vinyl abundances of related species. In Orion KL, we found the following values: N(CH3OH)/N(CH3CH2OH) ≃ 30 (Kolesniková et al. 2014); N(CH3OCOH)/N(CH3CH2OCOH) ≃ 18 (only considering the compact ridge components for the methyl formate column density given in Margulès et al. (2010); the column density of ethyl formate is that given in Tercero et al. 2013); N(CH3CN)/N(CH3CH2CN) ≃ 0.55 (only for the hot core components López et al. 2014), and N(CH3CH2CN)/N(CH2CHCN) ≃ 9 (López et al. 2014). We have to point out that these ratios are derived with data obtained with a single-dish telescope; therefore, a considerable uncertainty arises from the assignation of similar spatial distribution for these species. Moreover, the column density of a single-cloud component could be over/underestimated in the multi-component line profiles observed with the 30 m telescope. Nevertheless, it seems that the methyl species are more abundant than the ethyl species and, considering that only vinyl cyanide has been detected, the ethyl species could have larger abundances than the vinyl ones. Therefore, it is possible that ethyl acetate could be more abundant than vinyl acetate. Unfortunately, accurate frequencies are not available for ethyl acetate in the millimeter range.

5.2. ALMA SV data

In order to further explore the presence of vinyl acetate in Orion KL, we also searched for this species in the ALMA science verification (SV) data1 between 213.7 and 246.7 GHz. First, we determined the position where methyl acetate peaks. As expected, it is within the compact ridge component (α2000.0 = 05h35m14.117s, δ2000.0 = −05°22′36.54″). We extracted the ALMA SV averaged spectrum of 5 × 5 pixels (0.95″ × 0.95″) around this position. In addition to the higher sensitivity, one of the advantages of ALMA with respect to single-dish telescopes (to search for new molecules) is the drastic reduction of the line confusion limit. Whereas in the 30 m data we obtained the averaged spectrum over a beam diameter ranged from 30 to 9, the synthetic beam of the ALMA SV data is 1.90″ × 1.40″. Therefore, in the 30 m data, the spectrum is a mix of all molecules from all source components giving rise to a high level of blending of the lines. Taking the mentioned advantage, we could detect tentatively vinyl acetate towards the compact ridge position of Orion KL with the ALMA SV data (see Fig. 4). In Fig. 4, the blue model provides the physical and chemical parameters that fit the observational ALMA data. The assumed parameters for this model are a source size of 3” (better adapted to the ALMA synthetic beam), vLSR = 7 km s-1, Δv = 2 km s-1, and TK = 150 K. Using the MADEX code in LTE approximation, we obtained a column density (A+E species) of 1 × 1015 cm-2. The red model uses the same parameters but a column density of 2 × 1015 cm-2, above which we could start to say that lines were missing or mismatched. We consider that the blue model is better adapted to the data, so we only provide a tentative detection. Although many lines are below the detection limit for the blue model,

we found that several lines are above this limit and no missing lines are found. To verify this tentative detection, we performed a synthetic spectrum with the same physical conditions but with the IRAM 30 m observations in order to search again for this molecule in these data. The new modeled line intensities appear below the detection limit, because the beam dilution factor for the 30 m is much higher than the gain by the higher column density obtained by ALMA. Therefore, no missing lines were found in the single-dish data confirming the tentative detection provided above.

Owing to the physical conditions of Orion KL, the maximum line density appears at the 1 mm window, i.e., at the wavelengths of the available observations with ALMA. Therefore, the confusion limit, even for filtered observations of the compact ridge is enough to hide the weak lines of vinyl acetate. Observations with ALMA at 2 mm could give us the major opportunity to confirm the detection of vinyl acetate in Orion KL.


Acknowledgments

The research leading to these results has received funding from the European Research Council under the European Union’s Seventh Framework Programme (FP/2007-2013)/ERC-2013-SyG, Grant Agreement No. 610256 NANOCOSMOS, Ministerio de Ciencia e Innovación (Grants CTQ2010-19008, CTQ2013-40717-P, and Consolider-Ingenio 2010 CSD2009-00038 program “ASTROMOL”) and Junta de Castilla y León (Grants VA070A08 and VA175U13). I.K. would like to thank the French PCMI (Programme National de Physique Chimie du Milieu Interstellaire).

References

All Tables

Table 1

Molecular constants in the rho-axis system of vinyl acetate obtained from the global fit using program BELGI-Cs.

All Figures

thumbnail Fig. 1

The lowest energy conformer of vinyl acetate (sp, ap) in the principal axis system.

Open with DEXTER
In the text
thumbnail Fig. 2

a) Broadband chirped pulse Fourier transform microwave spectrum (CP-FTMW) from 6 to 18 GHz. b) A section of the millimeter wave spectrum of vinyl acetate showing the assignments of the A- and E- components for the ground torsional state transitions.

Open with DEXTER
In the text
thumbnail Fig. 3

Loomis-Wood-type plot obtained from the AABS programs package (Kisiel et al. 2005) showing the identification of the stripes corresponding to the sequences of the b-type R-branch transitions for , and 4. Each sequence has two components, A and E, due to the methyl internal rotation. Rotational transitions are lined up to the central frequencies νcent of the A-component transitions with the rotational quantum numbers indicated on the right side of the diagram. The frequency distance Δν from the central frequency is documented on the x-axis. Each A-E doublet going to the left from the leading one at Δν = 0 corresponds to increasing Ka and decreasing J transitions. The highlighted horizontal stripe represents the part of the spectrum documented in Fig. 2b. For J′′< 60, the degeneracies for the energy levels with Ka = 3 and Ka = 4 are lifted and two separate sequences of (the left one) and (the right one) transitions are recognizable for both A- and E-components.

Open with DEXTER
In the text
thumbnail Fig. 4

Lines of vinyl acetate, CH3COOCHCH2, towards the compact ridge of Orion KL. The black histogram is the averaged spectrum over 5 × 5 pixels around the methyl acetate peak (see text) of the ALMA SV data (see text for references); these data have a 0.5 MHz of spectral resolution. The thin red line is the synthetic spectrum assuming a column density of 2 × 1015 cm-2 for vinyl acetate (A+E species). The thin blue line is the model for vinyl acetate with a column density of 1 × 1015 cm-2 (A+E species); this model corresponds to the given column density for the tentative detection of vinyl acetate (see text). A vLSR of 7 km s-1 is assumed.

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.