Astron. Astrophys. 324, 177-184 (1997)

1. Introduction

The pre-main sequence Herbig Ae/Be stars are objects of intermediate masses (2-5 )(Herbig 1960, Strom et al. 1972, Finkenzeller & Mundt 1984, Finkenzeller & Jankovicz 1984). They show all signs of intense stellar activity and strong stellar winds (Praderie et al. 1982, Catala et al 1986, Catala & Kunasz 1987). Their position in the HR diagram indicates that they are in the radiative contraction towards the main- sequence (Iben 1965, Gilliland 1986), and should in principle not possess any outer convective zone; therefore, if the young stellar evolutionary theory is correct, the classical magnetic dynamo mechanism could not be responsible for the observed phenomena. Finding the origin of this paradoxical activity is a major concern for young stellar evolution.

Two major interpretations should be studied in detail: the excess of non-radiative energy feeding the activity is i) provided by the energy of internal rotation or ii) by the gravitational energy of a circumstellar accretion disk. This question has already been adressed by Böhm (1993) by using various approaches based on high-resolution spectroscopy of an extended sample of Herbig Ae/Be stars. One of the major results of this study was the observation of centered and symmetric [O I](1F) forbidden emission lines in the observed sample of stars, which strongly questions the existence of significant circumstellar accretion disks around these young stars (Böhm and Catala 1994, hereafter "BC").

In a recent work Corcoran & Ray (1997) present new results on the shape and the centroids of forbidden emission line profiles in an extensive sample of 56 Herbig stars observed at medium and low resolution. 28 of these stars possess detectable [O I](1F) 6300.31 Å emission. Excluding 4 objects having a high velocity emission component in the [O I](1F) 6300.31 Å line, as well as 5 other objects measured at low precision, the statistical centroid distribution is in good agreement with the results announced in BC.

In some cases Herbig Ae/Be stars show the presence of jets and outflows (Mundt 1993, Goodrich 1993), some exhibit asymmetric forbidden lines (Corcoran & Ray 1994 and 1997, Hamann 1994) and some show elongated polarization maps (Bastien and Ménard 1990), indicating that in these particular cases of deeply embedded objects a "native" circumstellar disk might still have survived, while the majority of Herbig stars seems to have evolved out of their parental cloud towards a "naked" state, including the dissipation of their fossil/juvenile disk. The confirmation of this idea proposed for the first time in BC would directly indicate that we are observing Herbig stars in very different phases of their young stellar evolution; new support of this idea is found in Corcoran & Ray (1997), who exhibit a link between the degree of embeddedness and the amount of forbidden line blueshift.

However, the existence or absence of circumstellar accretion disks around Herbig Ae/Be stars is still matter of strong debate: the important infrared excesses observed around these stars can also be interpreted in two different ways. In the case of the lower mass pre-main sequence T Tauri stars the IR excess is generally attributed to the presence of accretion disks (Adams et al. 1987, Kenyon & Hartmann 1987, Bertout et al. 1988). It has therefore been appealing to build similar models for the Herbig Ae/Be stars (Hillenbrand et al. 1992, Lada & Adams 1992). The second approach attributes this excess to a circumstellar gas and dust halo of more or less spherical shape (Berrilli et al. 1992, Hartmann et al. 1993, Miroshnichenko et al. 1997).

It is straightforward that the absence or existence of circumstellar accretion disks is of major importance for our understanding of the young stellar evolution in general.

Due to their very low probability of radiative transition, the forbidden emission lines can only form in extremely low-density regions, therefore far away from the central star. The strongest forbidden emission lines in Herbig Ae/Be stars are the [O I](1F) lines. There are again two major explanations possible to explain the centered and symmetric emission profile (BC) observed in the spectra of this category of stars: i) the lines are formed in the remote parts of the stellar wind; the line formation regions are determined by the geometry of the stellar wind, which is most probably of spherical symmetry. In the case of the prototype Herbig Ae star AB Aur an estimate of a characteristical radial distance of the shell-like line formation region is of some AU (Böhm 1993). The symmetrical and centered profile of a line formed at these distances makes therefore the presence of an important circumstellar accretion disk impossible, since the receding part of the wind would be hidden by the opaque disk and yield a blueshifted and asymmetric emission profile. The only possibility to reconcile the shape of the observed emission profiles with the presence of an important disk would be ii) the formation of the forbidden emission line in a very thin and extended "disk atmosphere", following the surface of the disk. This atmosphere would just move in keplerian rotation together with the disk and would not present any significant orthogonal outflow velocity. This idea first proposed by Hirth et al. 1994a is based on the model of Kwan & Tademaru (1988), built for explaining the forbidden emission line profiles of the lower-mass T Tauri stars: this model contains two components, one at high velocity formed in a jet, and one "disk-atmosphere" component at almost zero radial velocity. If now, for some reason, the high-velocity component could not form, perhaps due to the different radiation field emanating from the hotter Herbig stars, the resulting profile would only be due to the low-velocity component and therefore be of symmetrical and centered appearance. However, since the emission volume of the forbidden lines has to be important due to the very low critical density of formation and the orthogonal extension of the "disk-atmosphere" can only be strongly limited, this would yield very high radial extensions of the emission region following the disk surface, most probably of the order of 100 and more AU (Böhm 1993).

All the above considerations about the shape and extent of the forbidden emission line regions around Herbig Ae/Be stars will have to be explained, in a subsequent step, by a complete physical model, explaining also the origin and the mechanism of the heating of these emission regions. At this step, only qualitative explanations can be advanced. In the case of a forbidden emission line formed in the outermost regions of the stellar wind the emission could occur in a post-shock region. This shock would then take place at a distance where the wind encounters the circumstellar gas and dust. The energy provided by the high velocity wind is largely sufficient to heat this region. In the case of a formation in a disk-atmosphere the limited gravitational energy of the accretion disk would have to satisfy the required energy balance.

In order to elucidate this interesting question, we therefore decided to select Herbig stars corresponding to different presumed evolutionary stages, and to perform on them long-slit spectroscopy in order to get first information about the spatial properties, i.e., about the shape and extend of the associated [O I](1F) forbidden emission line regions.

Sect. 2 presents the observations and data analysis. Results are presented in Sect. 3 and discussed in Sect. 4. Finally, the conclusions are drawn in Sect. 5.

© European Southern Observatory (ESO) 1997

Online publication: May 26, 1998

helpdesk.link@springer.de