Free Access
Issue
A&A
Volume 578, June 2015
Article Number A115
Number of page(s) 38
Section Interstellar and circumstellar matter
DOI https://doi.org/10.1051/0004-6361/201526270
Published online 15 June 2015

© ESO, 2015

1. Introduction

Low- and intermediate-mass stars (0.88 M) evolve onto the asymptotic giant branch (AGB) after exhaustion of the central He. During this phase, the stars continue nucleosynthesis in a thin shell of He, surrounded by a larger H-burning shell (e.g., Iben & Renzini 1983; Habing 1996). During its AGB lifetime, a star experiences a number of He-flashes that lead to a sudden increase in its luminosity over a brief period of time. Owing to efficient convection inside the star, the nucleosynthesis products are mixed outwards (the so-called third dredge-up). Fresh carbon produced in the He-shell burning is transported to the stellar photosphere and will increase the carbon-to-oxygen ratio, which can turn the originally oxygen-rich stars into carbon-rich stars.

The exact outcome of the nucleosynthesis and mixing events on the stellar surface abundance depends on the stellar mass. This is especially true when it comes to the process called hot-bottom burning, where the base of the convective zone of the hydrogen envelope is hot enough for CNO-cycle burning to destroy carbon (Lattanzio & Wood 2003). There is observational evidence that the process of hot-bottom burning occurs. Sackmann & Boothroyd (1992) observed a number of stars that exhibit anomalously high lithium abundance which is thought to be a byproduct of hot-bottom burning. This process is also thought to prevent a star from becoming carbon-rich. The minimum limit of the stellar mass that can trigger this process is estimated to be 5 M for stars with a solar metallicity and can be smaller for lower metallicities (Karakas & Lattanzio 2014). Another signpost that AGB stars are descendants of such intermediate-mass main-sequence stars is a low 12C/13C ratio – during the AGB phase, more massive intermediate-mass stars enter the CNO cycle which tends to produce 13C while converting 12C to 14N. The presence of 13C is the main neutron source for the s-process elements (Busso et al. 1999; Herwig 2005). Another possible route is thought to be from the 22Ne source in intermediate-mass stars, which requires a higher temperature.

An attempt to estimate masses of OH/IR stars near the galactic centre was carried out by Wood et al. (1998). They derived a mass of ~4 M for many of the stars in the sample using a period-luminosity relationship. Two stars with the longest periods were thought to have masses of up to 7 M. A number of optically visible OH/IR stars have been observed to exhibit a high Li abundance but show weak s-process elements (García-Hernández et al. 2007, 2013). These authors conclude that the stars have undergone hot-bottom burning but that there is a mechanism that delays the onset of s-process element production (Karakas et al. 2012). For OH/IR stars with an optically thick circumstellar envelope, which prevents the direct determination of stellar abundance, other hot-bottom burning indicators must be used. Two such stars (AFGL 5379 and OH 26.5+0.6) have been observed with the Herschel Space Observatory (hereafter Herschel, Pilbratt et al. 2010) to have strong water emission lines in H162\hbox{$_{2}^{16}$}O and H172\hbox{$_{2}^{17}$}O but no detection of the H182\hbox{$_{2}^{18}$}O line (Justtanont et al. 2013). An AGB evolutionary model for a 5 M star by Lattanzio & Wood (2003) shows that during hot-bottom burning, 18O is preferentially destroyed with respect to the other two isotopes. The Herschel observations of 18O/17O ratios of 1 for these two stars are in contrast to the value of ~3 derived from the interstellar medium (Wilson & Rood 1994).

In this paper, we present a larger sample of extreme OH/IR stars – those with very dusty circumstellar envelopes such that silicate dust features at 10 and 20 μm are in absorption, indicating high ( ≥ 10-4M yr-1) mass-loss rates. The observations of these stars were taken with all three instruments on board Herschel in order to search for the emission lines of three isotopologues of H2O as signposts for hot-bottom burning, with the aim to obtain a lower limit to the stellar mass. Detailed modelling of the line emission of H2O and other detected molecules will be presented in the future papers. In Sect. 2, we present the Herschel observations obtained as part of an open-time program on OH/IR stars. We discuss the results of our observations in Sect. 3 and summarize our findings in Sect. 4.

2. Observations

We obtained Herschel spectra of eight OH/IR stars selected from the sample based on Justtanont et al. (2006). These stars all exhibit the silicate dust features in absorption at both 10 and 18 μm and in some cases also show a water-ice band at 3.1 μm. Table 1 lists all stars observed and presented in this paper for the first time using the three Herschel instruments PACS, HIFI, and SPIRE. As noted, some stars have had their spectra taken as part of either a guaranteed time or another open-time program.

Table 1

OH/IR stars observed in the present work with observation identifiers (ObsID) indicated.

We observed four extreme OH/IR stars with the Herschel-HIFI instrument (de Graauw et al. 2010), which happened to be the last set of observations that Herschel did before the helium ran out at the end of April 2013 (Fig. 1). The frequency coverage for these stars are 1094.31098.4 GHz (lower side-band, LSB) and 1106.31110.4 GHz (upper side-band, USB). The LSB frequency covers the ortho-H2O transition of 312−303 for all three isotopologues while the USB permits the observations of the ground state transition of para-H172\hbox{$_{2}^{17}$}O 111−000. The raw data (level 0) were processed with pipeline version SPG 11.0 to obtain the level 1 and 2 data. Further data reduction was performed in HIPE121 (Ott 2010) with calibration files version HIFI_CAL_15. Both polarizations were averaged together and the final spectra had the baseline subtracted and rebinned. From Fig. 1 it can bee seen that all objects, with the exception of OH 30.7+0.4, show the emission line due to H162\hbox{$_{2}^{16}$}O. Using a routine in HIPE12, we converted the antenna temperature to a flux scale so that we have the same flux scale for all three instruments, assuming a point source. The line fluxes are listed in Table 2.

thumbnail Fig. 1

Herschel-HIFI observation of four extreme OH/IR stars. The spectra have been corrected for the LSR velocity for each object and rebinned to ~2 km s-1 sampling. The vertical lines indicate the expected position of the H182\hbox{$_{2}^{18}$}O, H172\hbox{$_{2}^{17}$}O, and H162\hbox{$_{2}^{16}$}O 312−303 transitions, from left to right. The expected position of the H172\hbox{$_{2}^{17}$}O 111−000 transition (dashed line) from the upper side-band can be seen at the far right.

Table 2

Line fluxes of isotopologues of H2O in the observed HIFI range.

We obtained SPIRE spectra of eight extreme OH/IR stars and included OH 26.5+0.6 from the archive. The resolution of these spectra is ~1.4 GHz, corresponding to 380 km s-1 at a frequency of 1097 GHz hence the lines are not resolved and each observed emission line can be a blend of different molecules. With the help of the HIFI spectra of OH/IR stars (Justtanont et al. 2012), we can resolve the contribution of strong emission lines mainly due to H2O and CO.

The SPIRE data were reduced using the calibration files SPIRE_CAL_12_2. The spectra suffer from high backgrounds due to the location of these objects in the galactic plane and most stars are thought to have a typical distance larger than 1 kpc. Interstellar emission lines due to [N ii] at 205.178 μm and [C i] at 370.423 and 609.150 μm can be seen in the spectra, with the exception of AFGL 5379 which is thought to be closer than the other objects. We performed background subtraction from our spectra by investigating individual off-centre beams and selected those that have similar background levels as the central beam. However, the region between 300400 μm is badly affected so that it is difficult to recover background subtracted data.

A number of the stars observed with SPIRE have been observed using the PACS instrument. Together with the PACS data taken by the MESS guaranteed time program (Groenewegen et al. 2011) for AFGL 5379 and from another open-time program (PI. M.J. Barlow) for OH 26.5+0.6, we have full spectral coverage from 50 to 670 μm for all the stars in our sample. For OH 127.8+0.0, the data were taken as part of the calibration time (Lombaert et al. 2013). We used the calibration files PACS_CAL_48_0 for our targets. The flux from the central 3 × 3 spaxels were extracted. We note that the data suffer from significant leakage in the red part of the spectrum, which means that the data between 95100 μm and beyond ~190 μm cannot be recovered.

In the PACS spectral range, we identify the emission lines as coming from H2O, and CO plus three sets of lines due to OH at 79, 119, and 163 μm (Fig. 4 and Appendix C). These lines have previously been reported by Sylvester et al. (1997) and Lombaert et al. (2013) and are thought to be the pumping line for the OH masers seen in these objects. Although the archived spectra from the short-wavelength spectrometer (LWS, Clegg et al. 1996) aboard the Infrared Space Observatory (ISO, Kessler et al. 1996) of some of these stars are very noisy, there may be a possible hint of an absorption of the infrared pumping line at 53 μm. The other infrared pumping line at 34.6 μm was first reported towards two supergiants (Justtanont et al. 1996; Sylvester et al. 1997), but has not been detected towards AGB stars observed with ISO. No strong emission lines due to other molecules apart from H2O, OH, and CO have been reported from OH 127.8+0.0 (Lombaert et al. 2013).

3. Discussion

Justtanont et al. (2013) found strong H162\hbox{$_{2}^{16}$}O and H172\hbox{$_{2}^{17}$}O emission lines in two extreme OH/IR stars, AFGL 5379 and OH 26.5+0.6, while there is no detectable emission due to H182\hbox{$_{2}^{18}$}O. This is in contrast to what is observed in the interstellar medium and the Sun (Wilson & Rood 1994) where 18O is more abundant than 17O. The question arises if other extreme OH/IR stars show the same behaviour.

3.1. Resolved HIFI spectra

In four objects, we obtained HIFI spectra that cover the frequency range of the 312−303 line for all three isotopologues. From Fig. 1, we detect the main line at 1097.365 GHz (273.200 μm) in all the sources except OH 30.7+0.4. The expansion velocity of these stars is typically 15 km s-1, hence with an observed velocity resolution of 1 km s1 the line is well resolved. The spectrum of OH 32.0-0.5 is too noisy to confirm the detection of H172\hbox{$_{2}^{17}$}O at 1096.414 GHz (273.430 μm); H172\hbox{$_{2}^{17}$}O can be seen in OH 30.1-0.7 and OH 32.8-0.3, along with the H172\hbox{$_{2}^{17}$}O 111−000 line from the upper side-band. No evidence of the H182\hbox{$_{2}^{18}$}O 312−303 line at 1095.627 GHz (273.626 μm) is seen in our spectra (Table 2). This is similar to the non-detection of H182\hbox{$_{2}^{18}$}O in AFGL 5379 and OH 26.5+0.6 where both H162\hbox{$_{2}^{16}$}O and H172\hbox{$_{2}^{17}$}O lines are clearly detected (Justtanont et al. 2013).

thumbnail Fig. 2

The continuum subtracted apodized SPIRE spectrum of AFGL 5379 (histogram) with the Gaussian fits for H2O (red), H172\hbox{$_{2}^{17}$}O (blue), CO (green), and H2S (yellow). Other molecules such as SiO, HCN, and the interstellar lines of [C i] and [N ii] are shown in cyan. The flux for SPIRE and PACS spectra is in W m-2μm-1.

thumbnail Fig. 3

The continuum subtracted apodized SPIRE spectrum of AFGL 5379 (histogram) with the Gaussian fits for H2O (red), H172\hbox{$_{2}^{17}$}O (blue), CO (green), and H2S (yellow). Other molecules are shown in cyan.

thumbnail Fig. 4

The continuum subtracted PACS spectrum of AFGL 5379 (histogram) corrected to the ISO-LWS flux level with the Gaussian fits for H2O (red) and H172\hbox{$_{2}^{17}$}O (blue). Other molecules are shown in cyan.

Based on these observations, we searched for the presence of all isotopologues of H2O in the SPIRE and PACS spectra.

3.2. SPIRE spectra

In these spectra, we are able to discern a number of possible H172\hbox{$_{2}^{17}$}O lines. The SPIRE spectrum of AFGL 5379 (Figs. 2 and 3) shows the identification of both isotopologues, along with other molecules such as CO, SiO, HCN, and H2S (see Appendix B) although the lines are not resolved owing to the poor spectral resolution of the SPIRE instrument (λ/ Δλ~370−1290 for 670194 μm). In order to calculate the line fluxes, we employed the special script in the HIPE data reduction package written for unapodized SPIRE data (SPIRE_linefitting.py), taking into account the fit to the sinc function of the line profiles. For display purposes, we show the apodized spectra (with the sinc function corrected) together with the calculated Gaussian line profiles because the lines are unresolved, with a width of 0.078 cm-1 for unapodized spectra, i.e. Fline = 1.08 × Fpeak × FWHM. However, one caveat of the derived line fluxes is that the decomposed molecular components depend only on the central frequencies of the lines and not on the expected transition line strengths. For this purpose we include species thath exhibit several transitions and select lines within the SPIRE range with an upper energy 1000 K. We did, however, add an exception for the H2O ν2 transitions as these lines can sometimes be bright when they are masing. We listed the pair of detected H162\hbox{$_{2}^{16}$}O and H172\hbox{$_{2}^{17}$}O fluxes in Table 3. The estimated uncertainty for unblended lines is 30%, increasing to ~50% for blended lines. For our purpose, blended lines are defined as lines that have two or more species that are separated by less than the spectral resolution of the instrument and so the peaks are indistinguishable. In many cases, there are lines that overlap with distinct peaks which makes the line flux determination cleaner than the blended lines. We note that a series of lines due to H2S are detected in the SPIRE wavelength range for most of our objects. The full list of all the lines detected and plots of the SPIRE spectra observed can be found in Appendix B and D, respectively.

3.3. PACS spectra

For the PACS spectra, we fitted a Gaussian to individual lines to derive a line flux as the lines are also unresolved with a resolving power of 10005000, corresponding to a velocity resolution of ~30060 km s-1, for the long and short wavelengths, respectively. Since most of the lines are due to H2O, we decided to fit these before attributing the unfitted lines to other molecules such as CO and H2S. Figure 4 shows the continuum subtracted spectrum of AFGL 5379. Unfortunately, the source was not at the central spaxel when it was observed. This resulted in a loss of flux. However, we corrected this based on the archived ISO-LWS continuum flux data. The PACS data have been multiplied by a factor of 1.78 to get the flux to agree with the ISO-LWS flux. Another artefact is that the shortest wavelength part of the PACS spectrum cannot be recovered as it is badly affected by the foreground and background interstellar [O i] 63 emission. We did not detect any emission lines in the PACS spectrum of OH 21.5+0.5. The PACS spectrum of OH 30.7+0.4 does not show definitive detections of H172\hbox{$_{2}^{17}$}O lines. These two objects are hence not listed in Table 4.

The line fluxes of H162\hbox{$_{2}^{16}$}O and H172\hbox{$_{2}^{17}$}O are listed in Table 4. The estimated uncertainty for each derived line flux depends on the errors in baseline subtraction and the rms noise of the spectrum which can affect the flux by ~30%. Two H172\hbox{$_{2}^{17}$}O lines at 57.74 and 67.51 μm are close in wavelength to much higher excited lines of H2O 817−726 and 1148−1139, with upper energy levels of 1270 K and 2652 K, respectively), so we do not expect the calculated line fluxes to be much affected. We note here that although Tables 3 and 4 list only pairs of detected H162\hbox{$_{2}^{16}$}O and H172\hbox{$_{2}^{17}$}O, many H162\hbox{$_{2}^{16}$}O lines with no accompanying less abundant isotopologue are detected. The full list of all the lines detected and plots of the PACS spectra observed can be found in Appendix C and E, respectively.

3.4. H2O isotopologues

thumbnail Fig. 5

Observations taken from the HIFISTARS sample showing the 312−303 transition of ortho-H2O (see Fig. 1). The H182\hbox{$_{2}^{18}$}O line is clearly detected in W Hya and IK Tau (Justtanont et al. 2012) which have much lower mass-loss rates than in AFGL 5379 and OH 26.5+0.6 (Justtanont et al. 2013).

In all cases, we also searched for the presence of H182\hbox{$_{2}^{18}$}O in the HIFI, SPIRE and PACS spectra. It is clear that in the HIFI spectra where line blending is not an issue, the 312−303 transition at 1095.627 GHz is below the noise limit (Fig. 1). In PACS and SPIRE spectra, portions containing isolated unblended transition with another possible lines have been carefully looked at, but there is no emission detected above the noise. With these results, we conclude that our sample of extreme OH/IR star spectra lacks the presence of H182\hbox{$_{2}^{18}$}O. The upper limit of the H182\hbox{$_{2}^{18}$}O/H172\hbox{$_{2}^{17}$}O line ratios are given in Appendix A. Here, it is clear that the ratios are below unity in cases where H172\hbox{$_{2}^{17}$}O lines are detected above the noise. In the sample of oxygen-rich AGB stars from the guaranteed time program HIFISTARS with a lower mass-loss rate, with the same frequency settings as for the OH/IR stars observed in our open-time program (Justtanont et al. 2012), the line fluxes of H182\hbox{$_{2}^{18}$}O are always brighter than those of H172\hbox{$_{2}^{17}$}O for both the 111−000 and 312−303 transitions (Fig. 5).

The absence of H182\hbox{$_{2}^{18}$}O in these stars is counter-intuitive considering the observed isotopic ratio of 18O/17O ~ 3 in the interstellar medium. However, calculations of nucleosynthesis during the AGB phase for intermediate-mass stars predict that for stars with an initial mass larger than 5 M, the temperature at the base of the convective layer is high enough to start hot-bottom burning, preventing the star from becoming a carbon star (see e.g., Lattanzio et al. 1996; Lattanzio & Wood 2003). At the beginning of hot-bottom burning, 18O is destroyed while the production of 17O is increased by an order of magnitude, hence the 18O/17O ratio has an expected value of 10-6 while the 16O/17O ratio is expected to be 350 (Lattanzio et al. 1996). Hot-bottom burning will finally cease when the star loses most of its mass such that the envelope mass is below 1 M. However, the third dredge-up can still continue and will change the 12C/13C ratio while leaving isotopic ratios of other elements almost unaffected. Studies of pre-solar grains reveal very few grains with extremely low 18O/16O ratios, which can possibly come from intermediate-mass AGB stars (Lugaro et al. 2007; Nittler et al. 2010), while most of the grains show oxygen isotopic ratios commonly expected from low-mass stars. The rarity of these presolar grains with low 18O content is consistent with the expected population of intermediate-mass stars assuming an initial mass function (Salpeter 1955; Scalo 1986) and the relatively short lifetimes of such stars.

The line flux ratios of H162\hbox{$_{2}^{16}$}O/H172\hbox{$_{2}^{17}$}O vary between about unity and less than 10 where the corresponding transition of both molecules is detected (Tables 3 and 4). Most of our reliable line flux ratios from HIFI give values between 2 and 5. This clearly indicates that at least the main line is optically thick. In order to derive isotopic abundance ratio from our observations, a radiative transfer calculation must be performed, which will be addressed in a forthcoming paper.

4. Summary

The H2O line fluxes observed with Herschel are presented for a sample of nine extreme OH/IR stars. These stars are close to the galactic plane and are thought to be population I stars. They all show strong H2O emission from the main isotopologue and from H172\hbox{$_{2}^{17}$}O. The absence of H182\hbox{$_{2}^{18}$}O detection was unexpected considering the solar and galactic ratio of 18O/17O of 35 (Wilson & Rood 1994; Wouterloot et al. 2008).

To explain this question, we propose that our sample stars have undergone hot-bottom burning, which preferentially destroys 18O relative to the other two isotopes. During hot-bottom burning, the abundance of 17O is expected to go up by an order of magnitude while the 18O abundance drops by more than two orders of magnitude. For such a process to happen, the bottom of the convective layer is required to be hotter than 80 × 106 K. This high temperature can be achieved in stars with initial masses of at least 5 M (Karakas & Lattanzio 2014). It should be noted that hot-bottom burning ceases when the star loses sufficient mass that the high temperature cannot be maintained. Although the isotopic ratios of most elements remain the same after this cessation, the 12C abundance can increase thanks to the continuation of the third dredge-up process bringing up carbon made by the triple-alpha reaction. The materials expelled from these stars will have an impact on local isotopic ratios and may also affect the overall chemical evolution of the Galaxy.

The Herschel observations of OH/IR stars complement previous optical ground based AGB star observations of the 7Li line by García-Hernández et al. (2013), which provides another indication of the operation of the hot-bottom burning in intermediate-mass stars. It may also be possible to search for signatures of hot-bottom burning using elements synthesized during this phase, such as 22Ne and 25Mg. Observations of isotopic ratios of various elements together with theoretical calculations of nucleosynthesis can yield better constraints on the initial mass of these stars.

Online material

Table 3

Line fluxes of isotopologues of H2O in the observed SPIRE range.

Table 4

Line fluxes of isotopologues of H2O in the observed PACS range.

Appendix A: Line flux ratios

This section of the appendices shows the line flux ratios of H182\hbox{$_{2}^{18}$}O/H172\hbox{$_{2}^{17}$}O. We note that since the H182\hbox{$_{2}^{18}$}O is not detected, we give lower limits to the line ratios. In the case when both isotopologues are not detected, the ratios are set to 1.0.

Table A.1

Line flux ratios of H182\hbox{$_{2}^{18}$}O/H172\hbox{$_{2}^{17}$}O observed by Herschel-HIFI.

Table A.2

Line flux ratios of H182\hbox{$_{2}^{18}$}O/H172\hbox{$_{2}^{17}$}O in the observed PACS range.

Table A.3

Ratios of H182\hbox{$_{2}^{18}$}O/H172\hbox{$_{2}^{17}$}O line fluxes observed with SPIRE.

Appendix B: Compilation of observed SPIRE line fluxes

We present a table with line fluxes for the SPIRE spectra of extreme OH/IR stars.

Table B.1

Line fluxes of all lines detected in the observed SPIRE range.

Appendix C: Compilation of observed PACS line fluxes

We present tables for the list of lines detected in the PACS spectra for individual objects.

Table C.1

Line fluxes calculated from PACS spectra of AFGL 5379.

Table C.2

Line fluxes calculated from PACS spectra of OH 26.5+0.6.

Table C.3

Line fluxes calculated from PACS spectra of OH 30.1-0.7.

Table C.4

Line fluxes calculated from PACS spectra of OH 32.0-0.5.

Table C.5

Line fluxes calculated from PACS spectra of OH 32.8-0.3.

Table C.6

Line fluxes calculated from PACS spectra of OH 42.3-0.1.

Appendix D: SPIRE spectra

This section shows the continuum subtracted apodized SPIRE spectra (in W m-2μm-1) of the stars in our sample (histogram) together with the Gaussian fits for H2O (red), H172\hbox{$_{2}^{17}$}O (blue), CO (green) and H2S (yellow). Other molecules are shown in cyan.

thumbnail Fig. D.1

The continuum subtracted apodized SPIRE spectrum of OH21.5+0.5.

thumbnail Fig. D.2

The continuum subtracted apodized SPIRE spectrum of OH21.5+0.5.

thumbnail Fig. D.3

The continuum subtracted apodized SPIRE spectrum of OH 127.8+0.0.

thumbnail Fig. D.4

The continuum subtracted apodized SPIRE spectrum of OH 127.8+0.0.

thumbnail Fig. D.5

The continuum subtracted apodized SPIRE spectrum of OH 26.5+0.6.

thumbnail Fig. D.6

The continuum subtracted apodized SPIRE spectrum of OH 26.5+0.6,

thumbnail Fig. D.7

The continuum subtracted apodized SPIRE spectrum of OH 30.7+0.4.

thumbnail Fig. D.8

The continuum subtracted apodized SPIRE spectrum of OH 30.7+0.4.

thumbnail Fig. D.9

The continuum subtracted apodized SPIRE spectrum of OH 30.1-0.7.

thumbnail Fig. D.10

The continuum subtracted apodized SPIRE spectrum of OH 30.1-0.7.

thumbnail Fig. D.11

The continuum subtracted apodized SPIRE spectrum of OH 32.0-0.5.

thumbnail Fig. D.12

The continuum subtracted apodized SPIRE spectrum of OH 32.0-0.5.

thumbnail Fig. D.13

The continuum subtracted apodized SPIRE spectrum of OH 32.8-0.3.

thumbnail Fig. D.14

The continuum subtracted apodized SPIRE spectrum of OH 32.8-0.3.

thumbnail Fig. D.15

The continuum subtracted apodized SPIRE spectrum of OH 42.3-0.1.

thumbnail Fig. D.16

The continuum subtracted apodized SPIRE spectrum of OH 42.3-0.1.

Appendix E: PACS spectra

This section shows the continuum subtracted PACS spectra (in W m-2μm-1) of the stars in our sample (histogram) together with the Gaussian fits for H2O (red), H172\hbox{$_{2}^{17}$}O (blue), and CO (green). Other molecules are shown in cyan. We note that no circumstellar emission lines are detected in the spectrum of OH 21.5+0.5. The PACS spectrum of OH 127.8+0.0 has already been publish by Lombaert et al. (2013) and is not presented here.

thumbnail Fig. E.1

The continuum subtracted PACS spectrum of OH 21.5+0.5.

thumbnail Fig. E.2

The continuum subtracted PACS spectrum of OH 26.5+0.6.

thumbnail Fig. E.3

The continuum subtracted PACS spectrum of OH 30.7+0.4.

thumbnail Fig. E.4

The continuum subtracted PACS spectrum of OH 30.1-0.7.

thumbnail Fig. E.5

The continuum subtracted PACS spectrum of OH 32.0-0.5.

thumbnail Fig. E.6

The continuum subtracted PACS spectrum of OH 32.8-0.3.

thumbnail Fig. E.7

The continuum subtracted PACS spectrum of OH 42.3-0.1.


1

HCSS/HSpot/HIPE is are joint developments by the Herschel Science Ground Segment Consortium, consisting of ESA, the NASA Herschel Science Center, and the HIFI, PACS, and SPIRE consortia.

Acknowledgments

This research is partly funded by the Swedish National Space Board. We also thank both the referee and the editor for further comments for improvement of this paper. HIFI has been designed and built by a consortium of institutes and university departments from across Europe, Canada and the United States under the leadership of SRON Netherlands Institute for Space Research, Groningen, The Netherlands and with major contributions from Germany, France and the US. Consortium members are: Canada: CSA, U.Waterloo; France: CESR, LAB, LERMA, IRAM; Germany: KOSMA, MPIfR, MPS; Ireland, NUI Maynooth; Italy: ASI, IFSI-INAF, Osservatorio Astrofisico di Arcetri-INAF; Netherlands: SRON, TUD; Poland: CAMK, CBK; Spain: Observatorio Astronómico Nacional (IGN), Centro de Astrobiología (CSIC-INTA). Sweden: Chalmers University of Technology MC2, RSS & GARD; Onsala Space Observatory; Swedish National Space Board, Stockholm University Stockholm Observatory; Switzerland: ETH Zurich, FHNW; USA: Caltech, JPL, NHSC. PACS has been developed by a consortium of institutes led by MPE (Germany) and including UVIE (Austria); KU Leuven, CSL, IMEC (Belgium); CEA, LAM (France); MPIA (Germany); INAF-IFSI/OAA/OAP/OAT, LENS, SISSA (Italy); IAC (Spain). This development has been supported by the funding agencies BMVIT (Austria), ESA-PRODEX (Belgium), CEA/CNES (France), DLR (Germany), ASI/INAF (Italy), and CICYT/MCYT (Spain). SPIRE has been developed by a consortium of institutes led by Cardiff University (UK) and including Univ. Lethbridge (Canada); NAOC (China); CEA, LAM (France); IFSI, Univ. Padua (Italy); IAC (Spain); Stockholm Observatory (Sweden); Imperial College London, RAL, UCL-MSSL, UKATC, Univ. Sussex (UK); and Caltech, JPL, NHSC, Univ. Colorado (USA). This development has been supported by national funding agencies: CSA (Canada); NAOC (China); CEA, CNES, CNRS (France); ASI (Italy); MCINN (Spain); SNSB (Sweden); STFC and UKSA (UK); and NASA (USA).

References

  1. Busso, M., Gallino, R., & Wasserburg, G. J. 1999, ARA&A, 37, 239 [NASA ADS] [CrossRef] [Google Scholar]
  2. Clegg, P. E., Ade, P. A. R., Armand, C., et al. 1996, A&A, 315, L38 [NASA ADS] [Google Scholar]
  3. de Graauw, T., Helmich, F. P., Phillips, T. G., et al. 2010, A&A, 518, L6 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  4. García-Hernández, D. A., García-Lario, P., Plez, B., et al. 2007, A&A, 462, 711 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. García-Hernández, D. A., Zamora, O., Yagüe, A., et al. 2013, A&A, 555, L3 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Groenewegen, M. A. T., Waelkens, C., Barlow, M. J., et al. 2011, A&A, 526, A162 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  7. Habing, H. J. 1996, A&ARv, 7, 97 [NASA ADS] [CrossRef] [Google Scholar]
  8. Herwig, F. 2005, ARA&A, 43, 435 [NASA ADS] [CrossRef] [Google Scholar]
  9. Iben, Jr., I., & Renzini, A. 1983, ARA&A, 21, 271 [NASA ADS] [CrossRef] [Google Scholar]
  10. Justtanont, K., Skinner, C. J., Tielens, A. G. G. M., Meixner, M., & Baas, F. 1996, ApJ, 456, 337 [NASA ADS] [CrossRef] [Google Scholar]
  11. Justtanont, K., Olofsson, G., Dijkstra, C., & Meyer, A. W. 2006, A&A, 450, 1051 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  12. Justtanont, K., Khouri, T., Maercker, M., et al. 2012, A&A, 537, A144 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  13. Justtanont, K., Teyssier, D., Barlow, M. J., et al. 2013, A&A, 556, A101 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  14. Karakas, A. I., & Lattanzio, J. C. 2014, PASA, 31, 30 [Google Scholar]
  15. Karakas, A. I., García-Hernández, D. A., & Lugaro, M. 2012, ApJ, 751, 8 [NASA ADS] [CrossRef] [Google Scholar]
  16. Kessler, M. F., Steinz, J. A., Anderegg, M. E., et al. 1996, A&A, 315, L27 [NASA ADS] [Google Scholar]
  17. Lattanzio, J., & Wood, P. R. 2003, in Asymptotic giant branch stars (Berlin, New York: Springer), eds. H. J. Habing, & H. Olofsson, Astron. Astrophysics Lib., 23 [Google Scholar]
  18. Lattanzio, J., Frost, C., Cannon, R., & Wood, P. R. 1996, Mem. Soc. Astron. Italiana, 67, 729 [Google Scholar]
  19. Lombaert, R., Decin, L., de Koter, A., et al. 2013, A&A, 554, A142 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Lugaro, M., Karakas, A. I., Nittler, L. R., et al. 2007, A&A, 461, 657 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  21. Nittler, L. R., Gyngard, F., & Zinner, E. 2010, Meteorit. Planet. Sci. Suppl., 73, 5245 [NASA ADS] [Google Scholar]
  22. Ott, S. 2010, in Astronomical Data Analysis Software and Systems XIX, eds. Y. Mizumoto, K.-I. Morita, & M. Ohishi, ASP Conf. Ser., 434, 139 [Google Scholar]
  23. Pilbratt, G. L., Riedinger, J. R., Passvogel, T., et al. 2010, A&A, 518, L1 [CrossRef] [EDP Sciences] [Google Scholar]
  24. Sackmann, I.-J., & Boothroyd, A. I. 1992, ApJ, 392, L71 [NASA ADS] [CrossRef] [Google Scholar]
  25. Salpeter, E. E. 1955, ApJ, 121, 161 [Google Scholar]
  26. Scalo, J. M. 1986, Fund. Cosmic Phys., 11, 1 [NASA ADS] [EDP Sciences] [Google Scholar]
  27. Sylvester, R. J., Barlow, M. J., Nguyen-Q-Rieu, et al. 1997, MNRAS, 291, L42 [NASA ADS] [CrossRef] [Google Scholar]
  28. Wilson, T. L., & Rood, R. 1994, ARA&A, 32, 191 [NASA ADS] [CrossRef] [Google Scholar]
  29. Wood, P. R., Habing, H. J., & McGregor, P. J. 1998, A&A, 336, 925 [NASA ADS] [Google Scholar]
  30. Wouterloot, J. G. A., Henkel, C., Brand, J., & Davis, G. R. 2008, A&A, 487, 237 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]

All Tables

Table 1

OH/IR stars observed in the present work with observation identifiers (ObsID) indicated.

Table 2

Line fluxes of isotopologues of H2O in the observed HIFI range.

Table 3

Line fluxes of isotopologues of H2O in the observed SPIRE range.

Table 4

Line fluxes of isotopologues of H2O in the observed PACS range.

Table A.1

Line flux ratios of H182\hbox{$_{2}^{18}$}O/H172\hbox{$_{2}^{17}$}O observed by Herschel-HIFI.

Table A.2

Line flux ratios of H182\hbox{$_{2}^{18}$}O/H172\hbox{$_{2}^{17}$}O in the observed PACS range.

Table A.3

Ratios of H182\hbox{$_{2}^{18}$}O/H172\hbox{$_{2}^{17}$}O line fluxes observed with SPIRE.

Table B.1

Line fluxes of all lines detected in the observed SPIRE range.

Table C.1

Line fluxes calculated from PACS spectra of AFGL 5379.

Table C.2

Line fluxes calculated from PACS spectra of OH 26.5+0.6.

Table C.3

Line fluxes calculated from PACS spectra of OH 30.1-0.7.

Table C.4

Line fluxes calculated from PACS spectra of OH 32.0-0.5.

Table C.5

Line fluxes calculated from PACS spectra of OH 32.8-0.3.

Table C.6

Line fluxes calculated from PACS spectra of OH 42.3-0.1.

All Figures

thumbnail Fig. 1

Herschel-HIFI observation of four extreme OH/IR stars. The spectra have been corrected for the LSR velocity for each object and rebinned to ~2 km s-1 sampling. The vertical lines indicate the expected position of the H182\hbox{$_{2}^{18}$}O, H172\hbox{$_{2}^{17}$}O, and H162\hbox{$_{2}^{16}$}O 312−303 transitions, from left to right. The expected position of the H172\hbox{$_{2}^{17}$}O 111−000 transition (dashed line) from the upper side-band can be seen at the far right.

In the text
thumbnail Fig. 2

The continuum subtracted apodized SPIRE spectrum of AFGL 5379 (histogram) with the Gaussian fits for H2O (red), H172\hbox{$_{2}^{17}$}O (blue), CO (green), and H2S (yellow). Other molecules such as SiO, HCN, and the interstellar lines of [C i] and [N ii] are shown in cyan. The flux for SPIRE and PACS spectra is in W m-2μm-1.

In the text
thumbnail Fig. 3

The continuum subtracted apodized SPIRE spectrum of AFGL 5379 (histogram) with the Gaussian fits for H2O (red), H172\hbox{$_{2}^{17}$}O (blue), CO (green), and H2S (yellow). Other molecules are shown in cyan.

In the text
thumbnail Fig. 4

The continuum subtracted PACS spectrum of AFGL 5379 (histogram) corrected to the ISO-LWS flux level with the Gaussian fits for H2O (red) and H172\hbox{$_{2}^{17}$}O (blue). Other molecules are shown in cyan.

In the text
thumbnail Fig. 5

Observations taken from the HIFISTARS sample showing the 312−303 transition of ortho-H2O (see Fig. 1). The H182\hbox{$_{2}^{18}$}O line is clearly detected in W Hya and IK Tau (Justtanont et al. 2012) which have much lower mass-loss rates than in AFGL 5379 and OH 26.5+0.6 (Justtanont et al. 2013).

In the text
thumbnail Fig. D.1

The continuum subtracted apodized SPIRE spectrum of OH21.5+0.5.

In the text
thumbnail Fig. D.2

The continuum subtracted apodized SPIRE spectrum of OH21.5+0.5.

In the text
thumbnail Fig. D.3

The continuum subtracted apodized SPIRE spectrum of OH 127.8+0.0.

In the text
thumbnail Fig. D.4

The continuum subtracted apodized SPIRE spectrum of OH 127.8+0.0.

In the text
thumbnail Fig. D.5

The continuum subtracted apodized SPIRE spectrum of OH 26.5+0.6.

In the text
thumbnail Fig. D.6

The continuum subtracted apodized SPIRE spectrum of OH 26.5+0.6,

In the text
thumbnail Fig. D.7

The continuum subtracted apodized SPIRE spectrum of OH 30.7+0.4.

In the text
thumbnail Fig. D.8

The continuum subtracted apodized SPIRE spectrum of OH 30.7+0.4.

In the text
thumbnail Fig. D.9

The continuum subtracted apodized SPIRE spectrum of OH 30.1-0.7.

In the text
thumbnail Fig. D.10

The continuum subtracted apodized SPIRE spectrum of OH 30.1-0.7.

In the text
thumbnail Fig. D.11

The continuum subtracted apodized SPIRE spectrum of OH 32.0-0.5.

In the text
thumbnail Fig. D.12

The continuum subtracted apodized SPIRE spectrum of OH 32.0-0.5.

In the text
thumbnail Fig. D.13

The continuum subtracted apodized SPIRE spectrum of OH 32.8-0.3.

In the text
thumbnail Fig. D.14

The continuum subtracted apodized SPIRE spectrum of OH 32.8-0.3.

In the text
thumbnail Fig. D.15

The continuum subtracted apodized SPIRE spectrum of OH 42.3-0.1.

In the text
thumbnail Fig. D.16

The continuum subtracted apodized SPIRE spectrum of OH 42.3-0.1.

In the text
thumbnail Fig. E.1

The continuum subtracted PACS spectrum of OH 21.5+0.5.

In the text
thumbnail Fig. E.2

The continuum subtracted PACS spectrum of OH 26.5+0.6.

In the text
thumbnail Fig. E.3

The continuum subtracted PACS spectrum of OH 30.7+0.4.

In the text
thumbnail Fig. E.4

The continuum subtracted PACS spectrum of OH 30.1-0.7.

In the text
thumbnail Fig. E.5

The continuum subtracted PACS spectrum of OH 32.0-0.5.

In the text
thumbnail Fig. E.6

The continuum subtracted PACS spectrum of OH 32.8-0.3.

In the text
thumbnail Fig. E.7

The continuum subtracted PACS spectrum of OH 42.3-0.1.

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.