EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 541 (May 2012) A&A, 541 (2012) L11 Full HTML
Free Access
Issue
A&A
Volume 541, May 2012
Article Number L11
Number of page(s) 4
Section Letters
DOI https://doi.org/10.1051/0004-6361/201219379
Published online 22 May 2012

© ESO, 2012

1. Introduction

Water is an essential molecule in star-forming regions, in particular because it is the main constituent of the icy mantles of dust grains, one of the main repositories of oxygen, and in certain conditions an important gas coolant (e.g. Nisini et al. 2010). It is moreover detected in virtually all star-forming environments in which it is searched for (e.g. Caselli et al. 2010; Hogerheijde et al. 2011; van Dishoeck et al. 2011).

Water is believed to form efficiently on the surface of dust grains through several different pathways. The relative importance of these pathways is not completely understood, however, and the reaction barriers involved are still under debate despite numerous recent laboratory experiments (Ioppolo et al. 2010; Cuppen et al. 2010). This is because isolating a particular reaction in the laboratory is extremely difficult, and the absolute measure of the reaction rates and barriers on the grains is almost impossible.

In this sense, astronomical observations of the precursors of water can be of great help to constrain the chemical network that leads to water formation. A significant step in the understanding of water formation was achieved thanks to the detection with APEX of HOOH in Oph A (Bergman et al. 2011b). This detection has been carefully modeled with an astrochemical model combining reactions in the gas phase and on the grain surfaces based on the hybrid moment equation (HME) method (Du & Parise 2011), applied to the conditions of the Oph A cloud (Du et al. 2012). The model successfully reproduces simultaneously the abundances of the observed gas-phase HOOH, O2, H2CO and CH3OH. The model also predicts that the hydroperoxyl radical (HO2, or O2H), the direct precursor leading to HOOH via hydrogenation, has a significant abundance, which makes its detection with current radio telescopes possible. Hydroperoxyl has not yet – to our knowledge – been detected in the interstellar medium in either its solid or its gaseous form.

We aim here at detecting this radical toward the SM1 core of the ρ Oph A cloud, where HOOH and O2 (Liseau et al. 2012) were detected. The paper is organized as follows. The observations are presented in Sect. 2, and the results in Sect. 3. We discuss the implications of this detection in Sect. 4.

Table 1

The HO2 and SO2 lines detected in the same setups.

thumbnail Fig. 1

Energy level diagram of HO2 below 50 K. Only the a-type transitions listed in the JPL catalog were included and are drawn as arrows. Red arrows indicate observed transitions. The b-type transitions (connecting Ka = 0 and 1 states) were omitted for the sake of clarity. Next to each level the quantum number designation is listed.

thumbnail Fig. 2

Lines observed with the IRAM telescope. The red bars show the expected frequencies for the different lines, as measured in the laboratory (Saito 1977). The blue bars show the calculated frequencies from Charo & de Lucia (1982). The black vertical line in the lower panel shows the position of the undetected F = 1 → 0 line.

2. Observations

The HO2 radical is a light asymmetric rotor with an unpaired electronic spin. This causes each asymmetric rotor level (except 00,0) to be split into a doublet of states. This is illustrated in the HO2 energy level diagram in Fig. 1 where all states below 50 K are shown. It is based on the JPL catalog (Pickett et al. 1998), which uses mm and submm spectroscopy data from Beers & Howard (1975), Saito (1977), and Charo & de Lucia (1982). Also, the hyperfine splitting is large enough to be detectable. The spectrum of HO2 exhibits both a- and b-type transitions with associated dipole moments μa = 1.41 D and μb = 1.54 D (Saito & Matsumura 1980).

We targeted lines of the HO2 radical at 130.3 GHz and 260.6 GHz (see Table 1), which were observed in the laboratory by Saito (1977) and Charo & de Lucia (1982), respectively. We used the frequencies tabulated in the JPL database, which correspond to the measured values for these lines. The 10,1 → 00,0 lines around 65 GHz are poorly suited for a ground-based search because they are heavily affected by the strong atmospheric absorption due to O2. The 30,3 → 20,2 lines (at 195 GHz) fall in a frequency region where few receivers are currently in operation. They are also somewhat affected by the atmospheric absorption from the 183 GHz water line. Likewise, the 50,5 → 40,4 lines are affected by the atmospheric 325 GHz water line. Hence, the targeted lines at 130 and 260 GHz are the best candidates for a detection in a low-excitation source (for which most of the population is likely to reside in the lower Ka = 0 levels).

All observations targeted the SM1 position at α(2000) = 16h26m27.s\hbox{$\fs$}20, δ(2000) =  −24°24′04′′. The IRAM 30 m observations were carried out on Dec. 5, 2011, under average weather conditions with a precipitable water vapour (PWV) column of 5 mm. The EMIR receiver (Carter et al. 2012) was tuned in both polarizations to 130.3 GHz. The angular resolution of the observations is 19′′ at this frequency. The receiver was connected to FTS spectrometers, which were used in their highest resolution mode, providing a 50 kHz resolution. The focus was optimized on Saturn at the beginning of the observations. The pointing was checked locally every hour, either on 1741-038, 1622-297, or 1730-130, depending on the time of the pointing, and was found to be better than 3.″\hbox{$\farcs$}8 (rms). For the conversion to the Tmb scale, we adopt a main beam efficiency of 0.76, as extrapolated from the values given on the IRAM website1.

The APEX observations were carried out on Oct. 10 and 16, 2011, under variable weather conditions (PWV = 2.3–3.7 mm on Oct. 10, PWV = 0.6 mm on Oct. 16). The APEX1 receiver (Vassilev et al. 2008) was tuned to 260.25 GHz and connected to the XFFTS backend, leading to a velocity resolution of 0.09 km s-1. The angular resolution of the observations at this frequency is 24′′. Because no planet was available during the observations, the focus was optimized on IRAS 15194-5115 by maximizing the CO(2–1) line flux. The pointing was checked every hour by observing CO(2–1) on RAFGL1922, and was found to be always better than 3′′.

thumbnail Fig. 3

Lines observed with the APEX telescope. The data were smoothed to a resolution of 0.18 km s-1. The red bars show the expected frequencies for the different lines, as measured in the laboratory (Charo & de Lucia 1982). The blue bars show the computed hyperfine frequencies (see Table 1).

3. Results

Figure 2 displays the three lines targeted with the IRAM telescope, and Fig. 3 shows the two lines targeted with APEX. Over the wide bandwidth observed with the FTS, no obvious other line is present, except for two SO2 lines, which allow us to confirm the vlsr (3.8 km s-1) and line width of the expected lines (see Table 1).

We detected three HO2 lines with the IRAM telescope with a high signal-to-noise ratio, whose linewidths are consistent with the linewidth measured for the SO2 lines, but whose frequencies do not coincide exactly with those measured in the laboratory by Saito (1977), shown as red vertical bars in Fig. 2 (assuming a vlsr of 3.8 km s-1). Saito (1977) estimated an uncertainty on the laboratory frequency of 0.2 MHz (i.e. 0.5 km s-1 at 130  GHz). Charo & de Lucia (1982) measured the frequency of other lines between 150 and 550 GHz, and used the full dataset available at that time (including Saito’s lines) to fit the HO2 Hamiltonian. They found in particular a significant discrepancy (0.37 MHz) between the frequency measured by Saito (1977) and the calculated frequency from the Hamiltonian for the two lines at 130.26 GHz, which they excluded from their fit. The blue bars in Fig. 2 show the calculated frequency from Charo & de Lucia (1982). The first two detected lines coincide very closely with their theoretical frequencies. When adopting these new frequency values, the vlsr of the lines become 3.8 and 3.7 km s-1, respectively, which is fully consistent with the source velocity as determined by SO2 and several other molecules in the SM1 source (Bergman et al. 2011a). The third line is still slightly offset, with a vlsr of 3.4 km s-1.

Because the source is relatively line-poor, we can assign the three lines to HO2 with a high confidence level. An additional line (at 130.46368 GHz) is only tentatively detected, and we consider in the following only an upper limit on its flux. No blend with other plausible molecules was found at these frequencies in either the CDMS (Müller et al. 2001) or the JPL catalog (Pickett et al. 1998).

We detected two additional lines of HO2 with the APEX telescope. Both lines have an unresolved hyperfine structure. A Gaussian fit of these two lines leads to a broader line width than for the other lines (see Table 1), but this may be because of the hyperfine splitting, which for these lines corresponds to about 0.6–0.7 km s-1. Here again, no blend with any other plausible molecule is found.

thumbnail Fig. 4

HO2 rotational diagram. A source size of 24′′ is assumed. The triangles correspond to detected lines. The square corresponds to the upper limit of the flux on the undetected line at 130.464 GHz.

In summary, we report here the detection of five different spectral features assigned to seven different lines of HO2.

4. Analysis and discussion

We use the rotation diagram method to derive the abundance of HO2, which assumes that the population distribution can be described by a single temperature, and that lines are optically thin. The rotation diagram obtained with the five detected spectral features is presented in Fig. 4. We assume for lines 4 and 5 that the measured flux is coming from the two unresolved hyperfine components proportionally to Aul × gu, as expected for two optically thin lines originating from two levels of the same energy. We assume a source size of 24′′, as for hydrogen peroxide (Bergman et al. 2011b), and as derived by the analysis of Bergman et al. (2011a). We derive a rotational temperature of 16  ±  3 K, slightly lower than the rotational temperature derived for HOOH (22  ±  3 K, Bergman et al. 2011b), but marginally consistent with it within the error bars. The derived HO2 column density is (2.8  ±  1.0)  ×  1012 cm-2. This column density results in optically thin lines. Forcing the rotational temperature to be 22 K as for HOOH, we derive a column density for HO2 of 3.0  ×  1012 cm-2, very close to the column density derived with the fitted Trot of 16.4 K.

The relatively high Aul of the 260 GHz lines (and the even higher spontaneous rates of the b-type transitions) indicate that the excitation could be somewhat subthermal for the density in SM1 (105 − 106 cm-3, Bergman et al. 2011a). This could imply that our derived column density is too high. To estimate this, we simply truncate the partition function such that the summation only takes place over Ka = 0 states below 40 K and compare the result to the full partition function. We then find that the truncated part holds about 70% of the level populations at T = 16   K. Therefore, in the unlikely case of extreme subthermal excitation (no population in Ka > 0 states and in Ka = 0 states above 40,4), we may explain the observations by adopting a 30% lower HO2 column density. This lower column density falls within the errors and we conclude that subthermal excitation will have a minor impact on our derived column density.

From the H2CO and CH3OH analysis of the SM1 core, Bergman et al. (2011a) determined an H2 column density of 3  ×  1022 cm-2. The abundance of HO2 is therefore  ~10-10.

The gas-grain chemical model of Du et al. (2012), based on the detection of HOOH and other molecules, predicts that the HO2/HOOH abundance ratio in ρ Oph A should be about three. In this model, at T ~ 20 K, HO2 is mainly formed on dust grains through the barrierless reaction O + OH at an early stage (<5  ×  104 yr), and through the reaction H + O2 at a later stage. Here, we find observationally HO2/HOOH  ~  1, which is consistent with the model, taking into account the excitation uncertainty. This result is therefore an additional validation of our understanding of water formation on dust grains.

Motivated by our detection, we inspected previous spectral surveys for unidentified lines at the frequencies of the HO2 molecule. No line is detected in the TIMASSS survey (Caux et al. 2011) toward the low-mass protostar IRAS 16293−2422. Nummelin et al. (1998) reported (in their Table 46) two unidentified lines in their Sgr B2 survey that might correspond to HO2 frequencies, U260568 and U260771 toward Sgr B2(N) and U260569 toward Sgr B2(M). Because the lines have typical widths of 13 km s-1, line blending is a severe limiting factor for a secure line assignment. The Sgr B2 IRAM 30 m line survey coverage does not include the lines at 130 GHz (Belloche, priv. comm.). A 2 mm unpublished survey from B. Turner made with the NRAO 12 m telescope between 1993 and 1995 covers the two 130.26 GHz lines (but not the 130.46 GHz lines, Remijan et al. 2008). This doublet unfortunately falls very close to the J = 3–2 SiO emission/absorption feature at 130 268.7 MHz, making it very difficult to discern any weaker HO2 lines. Assuming that the unidentified lines of the Sgr B2(N) survey by Nummelin et al. (1998) are assigned to HO2, and assuming Trot = 50 K (Nummelin et al. 2000), we find N(HO2)  ≤  1.1  ×  1014 cm-2, averaged in the 20′′ beam. Using the H2 column density derived by Nummelin et al. (3  ×  1024 cm-2), this results in an upper limit on the fractional abundance of HO2 of 4  ×  10-11, somewhat below the detected abundance toward ρ Oph A.

The result of this work is one of the rare examples of the detection of a new molecule after it was predicted by chemical modeling (Du et al. 2012). Other examples include the detection of CF+ (Neufeld et al. 2006), motivated also by (gas-phase) model predictions. The novelty here is that the detection of HO2 was predicted by models describing the chemistry on dust grain surfaces. This shows the increasing predictive power of these models, thanks to the numerical refinements developed to correctly treat the stochasticity of surface chemistry (e.g. Garrod et al. 2009; Du & Parise 2011), and the recent numerous laboratory experiments (e.g. Oba et al. 2009; Ioppolo et al. 2010; Cuppen et al. 2010). The era when dust chemistry could be described as “the last refuge of the scoundrels” (Charnley et al. 1992) seems to be over.

Acknowledgments

We thank the referee, P. Goldsmith, for comments that helped improving the manuscript. We are grateful to the IRAM director, P. Cox, for granting us directorial time on the IRAM 30 m telescope. We thank A. Gusdorf for carrying out the APEX observations. B.P. and F.D. are supported by the German Deutsche Forschungsgemeinschaft, DFG Emmy Noether project number PA1692/1-1.

References

  1. Beers, Y., & Howard, C. J. 1975, J. Chem. Phys., 63, 4212 [NASA ADS] [CrossRef] [Google Scholar]
  2. Bergman, P., Parise, B., Liseau, R., & Larsson, B. 2011a, A&A, 527, A39 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  3. Bergman, P., Parise, B., Liseau, R., et al. 2011b, A&A, 531, L8 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  4. Carter, M., Lazareff, B., Maier, D., et al. 2012, A&A, 538, A89 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. Caselli, P., Keto, E., Pagani, L., et al. 2010, A&A, 521, L29 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Caux, E., Kahane, C., Castets, A., et al. 2011, A&A, 532, A23 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  7. Charnley, S. B., Tielens, A. G. G. M., & Millar, T. J. 1992, ApJ, 399, L71 [NASA ADS] [CrossRef] [Google Scholar]
  8. Charo, A., & de Lucia, F. C. 1982, J. Mol. Spect., 94, 426 [NASA ADS] [CrossRef] [Google Scholar]
  9. Cuppen, H. M., Ioppolo, S., Romanzin, C., & Linnartz, H. 2010, Phys. Chem. Chem. Phys. (Incorporating Faraday Transactions), 12, 12077 [NASA ADS] [CrossRef] [Google Scholar]
  10. Du, F., & Parise, B. 2011, A&A, 530, A131 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  11. Du, F., Parise, B., & Bergman, P. 2012, A&A, 538, A91 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  12. Garrod, R. T., Vasyunin, A. I., Semenov, D. A., Wiebe, D. S., & Henning, T. 2009, ApJ, 700, L43 [Google Scholar]
  13. Hogerheijde, M. R., Bergin, E. A., Brinch, C., et al. 2011, Science, 334, 338 [NASA ADS] [CrossRef] [PubMed] [Google Scholar]
  14. Ioppolo, S., Cuppen, H. M., Romanzin, C., van Dishoeck, E. F., & Linnartz, H. 2010, Phys. Chem. Chem. Phys. (Incorporating Faraday Transactions), 12, 12065 [NASA ADS] [CrossRef] [Google Scholar]
  15. Liseau, R., Goldsmith, P. F., Larsson, B., et al. 2012, A&A, 541, A73 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  16. Müller, H. S. P., Thorwirth, S., Roth, D. A., & Winnewisser, G. 2001, A&A, 370, L49 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  17. Neufeld, D. A., Schilke, P., Menten, K. M., et al. 2006, A&A, 454, L37 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  18. Nisini, B., Benedettini, M., Codella, C., et al. 2010, A&A, 518, L120 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  19. Nummelin, A., Bergman, P., Hjalmarson, Å., et al. 1998, ApJS, 117, 427 [NASA ADS] [CrossRef] [Google Scholar]
  20. Nummelin, A., Bergman, P., Hjalmarson, Å., et al. 2000, ApJS, 128, 213 [NASA ADS] [CrossRef] [Google Scholar]
  21. Oba, Y., Miyauchi, N., Hidaka, H., et al. 2009, ApJ, 701, 464 [NASA ADS] [CrossRef] [Google Scholar]
  22. Pickett, H. M., Poynter, R. L., Cohen, E. A., et al. 1998, J. Quant. Spec. Radiat. Transf., 60, 883 [Google Scholar]
  23. Remijan, A. J., Leigh, D. P., Markwick-Kemper, A. J., & Turner, B. E. 2008 [arXiv:0802.2273] [Google Scholar]
  24. Saito, S. 1977, J. Mol. Spect., 65, 229 [NASA ADS] [CrossRef] [Google Scholar]
  25. Saito, S., & Matsumura, C. 1980, J. Mol. Spect., 80, 34 [NASA ADS] [CrossRef] [Google Scholar]
  26. van Dishoeck, E. F., Kristensen, L. E., Benz, A. O., et al. 2011, PASP, 123, 138 [NASA ADS] [CrossRef] [Google Scholar]
  27. Vassilev, V., Meledin, D., Lapkin, I., et al. 2008, A&A, 490, 1157 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]

All Tables

Table 1

The HO2 and SO2 lines detected in the same setups.

In the text

All Figures

thumbnail Fig. 1

Energy level diagram of HO2 below 50 K. Only the a-type transitions listed in the JPL catalog were included and are drawn as arrows. Red arrows indicate observed transitions. The b-type transitions (connecting Ka = 0 and 1 states) were omitted for the sake of clarity. Next to each level the quantum number designation is listed.
In the text
thumbnail Fig. 2

Lines observed with the IRAM telescope. The red bars show the expected frequencies for the different lines, as measured in the laboratory (Saito 1977). The blue bars show the calculated frequencies from Charo & de Lucia (1982). The black vertical line in the lower panel shows the position of the undetected F = 1 → 0 line.

In the text
thumbnail Fig. 3

Lines observed with the APEX telescope. The data were smoothed to a resolution of 0.18 km s-1. The red bars show the expected frequencies for the different lines, as measured in the laboratory (Charo & de Lucia 1982). The blue bars show the computed hyperfine frequencies (see Table 1).

In the text
thumbnail Fig. 4

HO2 rotational diagram. A source size of 24′′ is assumed. The triangles correspond to detected lines. The square corresponds to the upper limit of the flux on the undetected line at 130.464 GHz.

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.