ForumSpringerAstron. Astrophys.
ForumWhats NewSearchOrders

Astron. Astrophys. 329, 161-168 (1998)

Previous Section Next Section Title Page Table of Contents

6. Discussion

In the following we investigate the evolutionary stage of the IR excess objects in M 17 by comparing them with other young stellar objects (YSOs).

6.1. The spectral index [FORMULA]

The spectral slope between 2.2 and 10 or 20 [FORMULA] m, defined as


is widely used to describe the evolution of YSOs. The observational evidence that [FORMULA] attains values between [FORMULA] for YSOs in the Ophiuchus dark cloud led to the following classification scheme (e.g. Lada 1987): Values of [FORMULA] are typical for the earliest stages of evolution where most of the observed luminosity is derived from accretion. The corresponding objects, referred to as Class I, are regarded as true protostars. Objects belonging to Class II ( [FORMULA] ) and Class III ( [FORMULA] ) are believed to be pre-main sequence stars surrounded by optically thick and optically thin circumstellar disks, respectively (André & Montmerle 1994). We note, that the steepest negative spectral index of -3.0 is always obtained when the observed wavebands fall into the Rayleigh-Jeans part of the blackbody emission.

In Fig. 3 we have calculated [FORMULA] between 2.2 and 20 [FORMULA] m (upper panel) for the emission of optically thin dust and a blackbody as a function of [FORMULA]. Values of [FORMULA] between 0 and +3 correspond either to dust radiation of 300 K [FORMULA] 600 K or to blackbody emission with slightly higher temperatures [FORMULA]. Obviously, the influence of extinction is not entirely negligible: [FORMULA] 40 mag may shift Class II objects of [FORMULA] 900 K into Class I. Moreover, all sources suffering [FORMULA] mag always appear as Class I objects; their detection at K, however, is rather unlikely. The influence of extinction becomes increasingly important when a spectral slope between 2.2 [FORMULA] m and wavelengths shorter than 10 [FORMULA] m is employed. The lower panel of Fig. 3shows an example for [FORMULA] between 2.2 and 5 [FORMULA] m: Already a visual extinction of [FORMULA] 30 mag converts any emission of arbitrary temperature into a Class I object, no matter whether the origin is pure stellar radiation or optically thin dust. Eventually, extinction becomes the dominating effect when deriving spectral indices e.g. from JHK data alone, thus making the classification of YSOs impossible.

[FIGURE] Fig. 3. Spectral indices [FORMULA] for the wavelengths interval 2.2 - 20 [FORMULA] m (upper panel) and 2.2 - 5 [FORMULA] m (lower panel) as a function of temperature and visual extinction [FORMULA] for the emission of interstellar dust and stellar radiation (solid: [FORMULA], dash-dotted: [FORMULA], dashed: [FORMULA] mag). The definition regions for Class I, II and III are separated by the horizontal dotted lines. The dominating influence of extinction with decreasing wavelength is demonstrated in the lower panel.

6.2. The evolutionary stage of the IR excess objects

Figs. 1and 2also show the location of some of the "cocoon stars " classified as such by CK2. Five of them (CEN 24, 31, 35, 43, 47) have data at B and V ; they all lie in the region of excess emission in Fig. 1. In Fig. 2the situation is less clear: While in the [FORMULA] -diagram no cocoon star displays abnormal colours, most of them develop an IR-excess with increasing wavelength. CEN 43 does not show a clear excess at longer wavelengths. The IR spectroscopy by Hanson & Conti (1995) classifies CEN 43 as a normal O type star, which does not exclude the presence of a dusty cocoon. According to its spectral index [FORMULA], which is 0.3 and 0.1 for 10 and 20 [FORMULA] m, respectively, CEN 43 might still be a Class I object. Only CEN 33 and 102 remain left of or close to the reddening line in all NIR-diagrams; both stars are classified as late type objects by Hanson & Conti (1995). In summary, this leaves six true "cocoon stars " in M 17 (CEN 24, 31, 34, 35, 43, 47) five of which have photometric data until 20 [FORMULA] m. Their SEDs fulfill the original condition [FORMULA] and thus qualifies them as genuine Class I objects. The empirical classification scheme for Class I suggested by Lada (1987) was also based on observations of five sources located in the Ophiuchus cloud (Lada & Wilking 1984, Elias 1978). Comparing both groups of Class I objects in M 17 and in Ophiuchus we find several differences:

i) The SEDs between 2.2 and 10 [FORMULA] m in Ophiuchus are steeper than those in M 17. Furthermore, all objects in Ophiuchus with [FORMULA] are "invisible " whereas those with [FORMULA] are "visible " (Lada 1987). Such a clear division does not exist in M 17: five out of the six "cocoon stars " are "visible ". These differences are probably all due to the amount of interstellar extinction: From the J and K fluxes Lada & Wilking (1984) estimated [FORMULA] values between 30 and 50 mag for the sources embedded in the Ophiuchus cloud. Employing the same technique and using the relation [FORMULA] as derived from Table 3 we obtain only [FORMULA] mag for the M 17 sources; both estimates assume that the intrinsic colours of the stars are negligible and that they have no significant excess emission in the [FORMULA] band. This comparison shows that the [FORMULA] of the Ophiuchus sources is more contaminated by extinction than that of the M 17 sources. As shown by Chini & Krügel (1985) only little circumstellar dust ( [FORMULA] mag) is necessary to produce large IR excesses at [FORMULA] [FORMULA] m. We therefore conclude that the steepness of the SEDs in Ophiuchus are to some extent the result of the large interstellar extinction whereas the SEDs in M 17 are intrinsically steeper, reflecting the dominant influence of circumstellar emission.


Table 3. Extinction law in M 17 and the diffuse ISM

ii) Integrating the observed luminosity [FORMULA] from 1.25 to 4.8 [FORMULA] m for both samples of YSOs we obtain an average value of [FORMULA] L [FORMULA] in Ophiuchus, excluding the object EL 29 which has 5 L [FORMULA]. The corresponding value for the "cocoon stars " is [FORMULA] L [FORMULA]. The observed luminosity [FORMULA] between 1.2 and 20 [FORMULA] m for the five Ophiuchus sources ranges between 0.8 and 14.8 L [FORMULA] with an average of [FORMULA] L [FORMULA] as derived from the data by Lada & Wilking (1984). The corresponding interval for the "cocoon stars " in M 17 is [FORMULA] with a mean of [FORMULA] L [FORMULA]. These estimates suggest that the Class I sources in M 17 are a factor of 60 more luminous than those in Ophiuchus.

After these remarks one may investigate the evolutionary status of the 20 new IR excess objects in M 17 by using the "cocoon stars " as template objects. Comparing the NIR properties of both groups, all newly identified IR excess objects lie significantly below the reddening lines in Fig. 2and are located in the same region as the six true "cocoon stars ". Their location is characterized by [FORMULA] and [FORMULA]. Because these relations are independent of the amount of extinction they directly measure the emission of circumstellar dust. As a consequence they are equivalent to the Class I criterion ( [FORMULA] ) and may be used as indicators for the evolutionary stage of an object in the absence of data longward of 4.8 [FORMULA] m.

The mean [FORMULA] from 1.25 to 4.8 [FORMULA] m for the 20 IR excess sources in M 17 is [FORMULA] and thus a factor of 60 higher than the corresponding value in the Ophiuchus cloud. This comparison shows again that the "cocoon stars " and the new IR excess objects belong to the same population of YSOs. They represent the youngest generation of early type stars in M 17 and are still enshrouded by the remnance of their protostellar cloud. The fact that these early stages of stellar evolution can be observed even at wavelengths below 2.2 [FORMULA] m is due to the fact that the ionizing radiation and the stellar winds from the nearby O type stars have partly cleared the region. We suggest, that these M 17 objects are the first high mass counterparts of classical Class I sources.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: November 24, 1997