Free Access
Issue
A&A
Volume 526, February 2011
Article Number A6
Number of page(s) 16
Section Catalogs and data
DOI https://doi.org/10.1051/0004-6361/200913984
Published online 13 December 2010

© ESO, 2010

1. Introduction

A planetary nebula is the most luminous transitory phase in the life of low and intermediate mass stars (0.6   M < M < 8   M) on their evolution from the asymptotic giant branch (AGB) to their final destiny, white dwarfs (WD). The PN phase begins once the central star reaches an effective temperature of 30 000 K and ionises the shell of material ejected during its evolution in the AGB. After about about 2 × 104 years, it ends when the nuclear burning in a thin shell of the star stops, and the nebula finally disperses.

PNe were discovered more than two centuries ago, and their number has increased every year, but there are still unsolved questions about them. Some of these, and perhaps the most important ones, are related to aspects of the central stars of the planetary nebulae (CSPN). Planetary nebulae nuclei are not located in a confined region of the HR diagram, and their optical spectra encompass all varieties known for hot stars, i.e. ranging from pure emission to emission-absorption mixtures and from near-continuous to pure strong absorption. The appearance of the spectrum depends upon temperature, luminosity, and chemical composition, or more fundamentally, upon core mass and state of evolution. Méndez (1991) suggested that the majority of CSPN can be classified in two distinct categories: those for which stellar H features can be identified in their spectra (hydrogen-rich) and those for which they cannot (hydrogen-poor).

At present, there are about 3000 confirmed and probable PNe known in our Milky Way, listed in Acker et al. (1992, 1996) (SECGPN1), (Parker et al. 2006; and Miszalski et al. 2002) (MASH2), and Drew et al. (2005) (IPHAS, INT Photometric H-Alpha Survey). However, spectroscopic information on their central stars is known only in a very small fraction of objects (about 13%, see Sect. 3).

Spectroscopy of CSPN is difficult to obtain because of their apparent low brightness, low apparent magnitudes (60% of the CSPN listed in the SECGPN have V > 15.5), and the surrounding gaseous shell whose emission lines often mask the stellar lines. In addition, the position of the CSPN is not always clear.

The determination of spectral types of CSPN should help significantly to improve our knowlege of their general evolutionary scheme, making it possible to consider CSPN as physical objects with individual parameters and peculiarities and not just as sources of ionizing radiation.

One of the first lists of CSPN was compiled by Aller (1948), then another was produced by Acker et al. (1982) (Catalogue of CSPN, Strasbourg Observatory). Information on CSPN can be found in the SECGPN and the MASH CDS-catalogues. Several authors have added contributions, although often for particular spectral types, e.g. WR+wels (Acker & Neiner 2003), B[e] (Lamers et al. 1998), evolved CSPN (Napiwotzki 1999), and PG 1159 (Werner & Herwig 2006).

To contribute to the knowledge of the final stellar evolution stages, we undertook a spectroscopic survey of CSPN and compiled a large list of CSPN. The motivation of the present work lies in a series of astronomical concerns: the complicated puzzle of different types of CSPN observed (see Table 1), few stars with spectral information, a lack of consensus in the evolutionary sequence of the CSPN, and the surprising bimodality in their hydrogen abundance.

This paper is organized as follows. The sample and observations are described in Sect. 2.1; in Sect. 2.2, we comment on the spectral classification; in Sect. 3, we present the catalogue of CSPN and we give a brief discussion. Finally, in Sect. 4 we present our conclusions.

Table 1

Summary of the spectral types of CSPN compiled in our catalogue, grouped by their atmospheric hydrogen abundance.

2. New spectral classification

2.1. Observations

We observed 45 southern CSPN selected from SECGPN and Boumis et al. (2003), the coordinates of which were taken from Kerber et al. (2003).

The observations were carried out during a three-year campaign between 2005 November and 2008 December that included a total of 31 nights of observations. For this survey, we used the REOSC spectrograph attached to the 2.15-m telescope at CASLEO, Argentina.

A 300 line mm-1 grating was used, which yielded a dispersion of 3.4 Å  pixel-1. During some nights, a grating of 600 line mm-1 was used (1.6 Å  pixel-1). The gratings provide a typical wavelength range of 3500−7000 Å (3875−5530 Å for the highest resolution). The slit was opened to 3′′ to be consistent with the seeing at the site.

2.2. Results

In this first work, we present a very preliminary classification of the observed CSPN. We distinguish between CSPN with absorption and emission lines. In the former group, we basically identified absorption lines of He i and He ii, these CSPN then being classified as OB. The latter group contained CSPN with identified emission lines, mainly of CIII (4650 Å and 5696 Å) and C iv (5806 Å), which are typical of [WC] stars. This CSPN were classified as “emission-line”. We obtained some spectra whose stellar continuum had a reasonable signal-to-noise ratio (S/N), but displayed, neither absorption nor emission lines. In these cases, although classified as “continuous” type, these objects are axpected to be H-rich (Kudritzki et al. 1981). Result are shown in Table 3. In a forthcoming paper, we perform a detailed spectroscopic analysis.

3. The catalogue of CSPN

3.1. Content

Taking into account that the information about CSPN spectral types is scattered among many publications, we carried out an extensive bibliographic compilation of the CSPN data with the goal of producing an updated list of those stars that have spectroscopic information. This list includes 492 stars of both confirmed and possible PN with spectral-type determinations, 45 of them from our own new data. Transition objects, such as post-AGB, PPN, or young-PN (Ej. V 348 Sgr, CRL 618, He 1-5, BD+33 2642, LS IV-12 111 and He 3-1475) were not included.

Table 2

Summary of results of KS test applied to the sample of Galactic latitude.

thumbnail Fig. 1

Distributions in Galactic longitude and latitude of CSPN (of true and possible PN) that belong to H-rich, H-poor, and wels star groups. Note that H-poor PN are more concentrated towards the Galactic center than H-rich ones. The similarity between the wels and H-poor distributions is also noticeable.

Open with DEXTER

The information included in the catalogue, discriminated between being confirmed and possible PN (Table 4), is:

  • Col. 1

    the PN G designation, taken from SECGPN;

  • Col. 2

    the common name of the object;

  • Cols. 3–4

    the equatorial coordinates (J2000.0) of the nebula, since in most cases there is no information on the position of the CSPN. Though in many cases this is evident, in others it is not;

  • Col. 5

    the spectral classification of the CSPN. If there are more than one, they are separated by a semi-colon (idem in the references column). However, we use only two spectral classifications if is it necessary, for example when the spectral classifications are very different. When the authors observed Balmer series absorption, we labeled these objects as H-rich. In some cases, the authors do not give the spectral type of the CS, but describe the identified lines. We also include the CSPN classified by Miszalski et al. (2002) in the MASHII catalogue: blue, [WR] or wels. Note that the blue characteristics of the CSPN images in MASHII is not based on any spectroscopic study;

  • Col. 6

    the reference where the spectral type was found (t.w. means this work);

  • Col. 7

    the reference that indicates whether the star is part of a binary system (nothing if not). Although some CSPN are of a late MK spectral type, it is accepted that the excitation source of the PN (if star and nebulae are physically associated) is a hitherto undetected hot star (Lutz 1977). In those cases, we include the label bc-CSPN, corresponding to binarity for the cool CSPN.

The catalogue of Acker et al. (1992) and AN03 provided spectroscopic information for 240 CSPN; with this new collated list, the number of CSPN with spectral classification has doubled. We hope that this new list will be useful for future investigations. In addition, we note that Parker et al. (2006) estimated that  ~30% of the MASH entries have candidate CSPN, with about half of these being high quality candidates suitable for immediate follow-up, so the list of CSPN with spectral classification will be increased quickly.

Table 3

Spectral types from our observations.

3.2. Discussion

The larger sample of CSPN with spectral types allows us to discuss the dichotomy between H rich and poor stars.

We grouped the H-rich and H-poor CSPN in Table 13, It is clear that the former group is more numerous than the H-poor one, the ratio being 1.4. In an earlier study, Méndez (1991) reported a ratio of 3. It is evident that stars with strong emission lines are easier to detect than those with absorption lines, thus favouring the detection of H-poor stars. However, is this effect strong enough to explain the ratio of stars observed between both groups?

We have found above that 30% of the whole CSPNe population appears to be hydrogen deficient (without counting the “blue” stars). It is difficult to obtain a theoretical prediction of this ratio of stellar types because the mechanism for generating H-poor CSPNe is not well known. The more accepted hypothesis for explaining the lack of hydrogen in the atmospheres of CSPN is the born-again phenomenon (Iben et al. 1983). In this framework, it is estimated that roughly 15% (Lawlor & MacDonald 2001) of post-AGB stars suffer a born-again event. Blöcker et al. (2001), based on their improved born-again models (thermal pulses plus overshooting), found that 20−25% of stars can be expected to become H-poor. These theoretical values are substantially lower than our observational value. According to this catalogue, it is difficult to imagine how a selection effect could be as efficient as to produce this high fraction of H-poor stars, so perhaps the born-again phenomenon is not the unique mechanism for obtaining an atmosphere free of hydrogen. We recall other ways to form H-poor CSPN, such as the binary channel (Tylenda & Górny 1993) or the continuous stripping of the outer H-rich layers by intense stellar winds (Górny & Tylenda 2000).

Only 71 close binary CSPN have been found (de Marco 2009; Miszalski et al. 2009b, and 2010), almost all of which have a H-rich spectra. The first [WR] star, in a close binary system, has been discovered in 2010 (Hajduk et al. 2010). We note that nearly 14% of the compiled CSPN are probably binary systems, in good agreement with the 10−15 value obtained by Bond et al. (1989).

We analyzed the distribution in Galactic coordinates of the CSPN sample that belongs to the H-rich and H-poor groups. From Fig. 1, it is evident that there is a strong concentration of H-poor and wels stars toward the Galactic center. This effect was observed by Górny et al. (2004) and attributed to a possible selection effect. However, it might be caused by the influence of metallicity in the mechanism that leads to an unleashing of the total hydrogen loss from the stellar atmosphere of those objects.

On the other hand, the average height above the Galactic plane of H-rich, H-poor and wels stars was found to be 13.9° ± 15.2, 9.0° ± 12.6, and 6.7° ± 5.3, respectively. As these errors are too large, we performed a Kolmogorov-Smirnov (KS) statistical analysis. The significance of the trends in KS test is assessed on the basis of differences, D, between their cumulative number distributions. This is used to define a probability coefficient P, such that low values of P imply significant differences. The results of the KS test are shown in Table 2. It is clear that the distribution of Galactic latitudes of H-rich and H-poor stars are very different. In addition, the sample of wels stars are, apparently, more similar to the H-poor stars than the other group, supporting the hypothesis that wels stars belong to the H-poor group and enhancing the ratio of H-poor to the whole CSPN population.

4. Conclusions

We have carried out a spectroscopic survey of PNe, during which we have performed a very preliminary determination of the spectral types of 45 of their central stars, all of them previously unclassified. In addition, we have performed an extensive bibliographic compilation of CSPN with determined spectral types. We have presented the list of 492 CSPN with spectral classification (together with their respective references), and included a tag indicating those that are either binary systems or candidates. We hope that this list will be useful for future investigations.

From our catalogue, we grouped CSPN whose atmospheres are hydrogen rich or poor; conservatively we ruled out the wels (nevertheless we found evidence supporting the hypothesis that wels belong to the H-poor group). We found that the ratio of stars in both groups is lower than previous estimates. According to our statistical analysis, we have found that PN with H-poor central star are more concentrated toward the Galactic center and Galactic plane than the H-rich group. This suggests that H-poor stars may have a more massive progenitor and in addition, the metallicity could play an important role in the mechanism responsible for generating hydrogen-free atmospheres. In addition, we have found that the frequency of occurrence of known close binaries among CSPNe is  ~14%.

Table 4

Catalogue of CSPN (true PN).

Table 5

Catalogue of CSPN (possible PN).

References to Tables 4 and 5:

  • A1966 - Abell (1966);

  • AK1985 - Aller & Keyes (1985);

  • AK1987 - Aller & Keyes (1987);

  • AN2003 - Acker & Neiner (2003);

  • CB2008 - Bilikova et al. (2008);

  • BC1999 - Bond & Ciardullo (1999);

  • BC2003 - Benetti et al. (2003);

  • B2008 - Bohigas (2008);

  • BD1993 - Bianchi & Defrancesco (1993);

  • BG1987 - Bond & Grauer (1987);

  • BL1990 - Bond & Livio (1990);

  • BM1993 - Bond et al. (1993);

  • BM2000 - Belczyski ă et al. (2000);

  • BO2002 - Bond et al. (2002);

  • BP2002 - Bond & Pollacco (2002);

  • BP2003 - Bond et al. (2003);

  • CB1999 - Ciardullo et al. (1999);

  • C1980 - Chromey (1980);

  • C1995 - Corradi (1995);

  • CB1999 - Ciardullo et al. (1999);

  • CG2009 - Chu et al. (2009);

  • CJ1987 - Cohen & Jones (1987);

  • CP1985 - Cerruti-Sola & Perinotto (1985);

  • D1983 - Drilling (1983);

  • D1985 - Drilling (1985);

  • D1999 - Dreizler (1999);

  • d2006 - de Marco (2006);

  • d2009 - de Marco (2009);

  • DB1996 - Duerbeck & Benetti (1996);

  • dS2003 - de Marco et al. (2003);

  • EP2005 - Exter et al. (2005);

  • F1999 - Feibelman (1999);

  • F1994 - Feibelman (1994);

  • FK1983 - Feibelman & Kaler (1983);

  • FM1981 - Ferguson et al. (1981);

  • FP2006 - Frew et al. (2006);

  • FS2010 - Frew et al. (2010);

  • GB1983 - Grauer & Bond (1983);

  • GB1987 - Grauer et al. (1987);

  • GC2009 - Górny et al. (2009);

  • GP2001 - Gauba et al. (2001);

  • GS2003 - Górny & Sidmiak (2003);

  • GS2004 - Górny et al. (2004);

  • GT2000 - Górny & Tylenda (2000);

  • GZ2003 - Gesicki & Zijlstra (2003);

  • GZ2006 - Gesicki et al. (2006);

  • HB2006 - Hillwig (2006);

  • HI2006 - Hsia et al. (2006);

  • HZ2010 - Hajduk et al. (2010);

  • H2003 - Handler (2003);

  • HD1984 - Heber & Drilling (1984);

  • HI2004 - Hewett & Irwin (2004);

  • HP1993 - Harrington & Paltoglou (1993);

  • HP2007 - Hultzsch et al. (2007);

  • JE1969 - Jones et al. (1969);

  • K1994 - Kondrat’eva (1994);

  • KB1994 - Kingsburgh & Barlow (1994);

  • KB2005 - Kraus et al. (2005);

  • KC1989 - Kwitter et al. (1989);

  • L1977 - Lutz (1977);

  • LK1987 - Lutz & Kaler (1987);

  • LS2000 - Liu et al. (2000);

  • LS2007 - Lee et al. (2007);

  • LR1983 - Law & Ritter (2001)

  • LZ1998 - Lamers et al. (1998);

  • MA2003 - Marcolino & derajoă (2003);

  • MA2009 - Miszalski et al. (2009a);

  • MA2010 - Miszalski (2010);

  • MASH-I - Parker et al. (2006);

  • MASH-II - Miszalski et al. (2008);

  • MC2006 - Mampaso et al. (2006);

  • MC2010 - Miszalski et al. (2010);

  • MD1981 - Margon et al. (1981);

  • MK1988 - Méndez et al. (1988a);

  • MN1977 - Méndez & Niemela (1977);

  • MN1982 - Méndez & Niemela (1982);

  • MN1981 - Méndez & Niemela (1981);

  • MO2007 - Mitchell et al. (2007a);

  • MP2001 - Morgan et al. (2001);

  • MP2007 - Mitchell et al. (2007b);

  • MV1997 - Miranda et al. (1997);

  • MV2010 - Miranda et al. (2010);

  • N1999 - Napiwotzki (1999);

  • NS1995 - Napiwotzki & Schoenberner (1995);

  • NT2005 - Napiwotzki et al. (2005);

  • P1983 - Pottasch (1983);

  • P1996 - Pottasch (1996);

  • P2004 - Pereira (2004);

  • PF2004 - Pierce et al. (2004);

  • PM2002 - Pea & Medina (2002);

  • PM2003 - Parker & Morgan (2003);

  • PM2008 - Pereira et al. (2008);

  • PR1997 - Pena et al. (1997);

  • PT1992 - Pena et al. (1992);

  • RC2001 - Rodrŋguez et al. (2001);

  • RD1998 - Rauch et al. (1998);

  • RH2002 - Rauch et al. (2002);

  • RK1999 - Rauch et al. (1999);

  • SB2007 - Smith et al. (2007);

  • SECGPN - Acker et al. (1992);

  • SF1987 - Sabbadin et al. (1987);

  • SG2010 - Santander-García (2010);

  • SK1994 - Stanghellini et al. (1994);

  • SL2004 - Shen et al. (2004);

  • SW1997 - Saurer et al. (1997);

  • SZ1997 - Soker & Zucker (1997);

  • TA1993 - Tylenda et al. (1993);

  • TK1996 - Tweedy & Kwitter et al. (1996);

  • TN2004 - Tovmassian et al. (2004);

  • TS1987 - Tamura & Shaw (1987);

  • WG2008 - Weidmann et al. (2008);

  • WH2006 - Werner & Herwig (2006);

  • WK1997 - Weinberger et al. (1997);

  • WW1996 - Walsh & Walton (1996);

  • WO1994 - Wlodarczyk & Olszewski (1994);

  • ZP1990 - Zijlstra et al. (1990).


1

Strasbourg-ESO Catalogue of Galactic PN (SECGPN) http://vizier.u-strasbg.fr/viz-bin/VizieR Planetary_Nebulae V/84/cstar.

2

Macquarie/AAO/Strasbourg Hα Planetary Galactic Catalog http://vizier.u-strasbg.fr/vizier/MASH

3

Although we have included the wels in the H-poor group (since we found evidence that wels and H-poor are in the same group), we prefer to be cautious and define and use the three groups H-rich, H-poor, and wels in the following discussion.

Acknowledgments

We would like to thank our anonymous referee whose critical remarks helped us to substantially improve this paper. The CCD and data acquisition system at CASLEO has been financed by R. M. Rich trough US NSF grant AST-90-15827. This work has been partially supported by Concejo de Investigaciones Cientifícas y Técnicas de la República Argentina (CONICET). This research has made use of the SIMBAD database, operated at CDS, Strasbourg, France. The authors wish to thank Drs. Guillermo Bosch and Roberto H. Méndez for comments that helped to improve the paper.

References

All Tables

Table 1

Summary of the spectral types of CSPN compiled in our catalogue, grouped by their atmospheric hydrogen abundance.

Table 2

Summary of results of KS test applied to the sample of Galactic latitude.

Table 3

Spectral types from our observations.

Table 4

Catalogue of CSPN (true PN).

Table 5

Catalogue of CSPN (possible PN).

All Figures

thumbnail Fig. 1

Distributions in Galactic longitude and latitude of CSPN (of true and possible PN) that belong to H-rich, H-poor, and wels star groups. Note that H-poor PN are more concentrated towards the Galactic center than H-rich ones. The similarity between the wels and H-poor distributions is also noticeable.

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.