ISOCAM Observations of extremely young Class 0 Candidates

H. Wiesemeyer ¹, P. Cox ², R. Guesten ³, & R. Zylka ⁴

¹ Institut de Radio Astronomie Millimetrique Domaine Universitaire de Grenoble 300, rue de la Piscine F-38406 St. Martin d'Heres CEDEX

² Institut d'Astrophysique Spatiale, Orsay, France

³ Max-Planck-Institut fr Radiaosatronomie, Bonn, Germany

⁴ Institut für Theoretische Astrophysik, Heidelberg, Germany

The transition between the ``Class -I'' and ``Class 0'' phases of protostellar evolution takes place between the formation of an observable, central accreting, hydrostatic object, and the verge of protostellar collapse and outflow activity characterizing the Class 0 phase (see e.g. Bontemps et al., 1996, A&A 311, 858). However, this evolutionary stage is difficult to distinguish from quiescent, starless cores (i.e. true ``Class -I'' protostars, or pre-stellar cores, see e.g. Ward-Thompson et al., 1998, ASP Conf.Ser. 132, 195, Boss & Yorke, ApJ 439, L55). The situation changes drastically as soon as the central hydrostatic object becomes observable. We successfully demonstrate that ISOCAM 6.75/15 micron observations are well suited to detect such objects, i.e. extremely young Class 0 protostars that still lack any important outflow activity. Statistically, this phase may be hard to observe, because the time span between the formation of an observable hydrostatic object and the onset of outflow activity may be rather short: if the initial density gradient is rather flat, the main accretion phase starts with a quite fast ``runaway'' collapse (Henriksen et al., 1997, A&A 323, 549).

There are basically two approaches to identify the transition phase: one consists of directly mapping the characteristic velocity field and density distribution of the protostellar envelopes, as seen in the rotational transition of a high-density molecular gas tracer. This approach is observationally challenging: it calls for high angular and velocity resolution, and a wide dynamic range across a field of size. It becomes only possible with the next generation of mm-interferometers.

The other approach consists of

- detecting a weak continuum emission from the central point source - or, alternatively, to evidence a cold, dense circumstellar cocoon by its absorption in front of a diffuse background) - to exclude as far as possible the presence of any well developped outflow activity,

- and to corroborate the core's contraction (not necessarily collapse !) by molecular spectra taken at high velocity and spatial resolution.

This contribution demonstrates that the hydrostatic objects at the center of extremely young protostellar envelopes are detectable with ISOCAM, even if they still have to accrete their major mass fraction. Among our six detections towards the L1082 filament (GF9 from the catalogue of globular filaments by Schneider & Elmegreen, 1979, ApJS 41, 87), we present the three most promising candidates. Together with complementary data, we confirm that these sources are still at the verge of, or in the early phase, of the main Class-0 phase. This conclusion is corroborated by

- narrow line profiles of high-density tracing molecular rotational transitions, that trace the contraction of the parent cloud core - the absence of any signpost of outflow activity (no CO linewings, no VLA 3.6 cm flux detected despite high sensitivity, no 2.2 micron emission) - extremely high luminosity ratios

>From our ISOCAM data, we can draw an evolutionary sequence: the youngest object is seen in absorption against the infrared background. It is most likely an externally heated core which central hydrostatic object - if present - still escapes detection. It only becomes visible in the ISOPHOT 180 micron image, but remains invisible at 90 micron (constraining its temperature to not more than 10 K). The second object is detected by ISOCAM, but highly reddened. Interferometer studies reveal that its envelope shows significant kinematic patterns. The third source also undergoes infall. It does neither drive an observable outflow, but has a core that was already detected by IRAS, though merely at 60 and 100 micron (where the emission may be rather dominated by the envelope).