4 Conclusions

Hard X-ray emission from a number of point sources is detected in a young, massive and embedded star-forming cluster in a very early stage of evolution. We did not detect X-ray emission from the most massive and central object (upper limit $L_{\rm {X}}<9\times 10^{34}$ erg s^-1 for kT=3 keV) but from a few sources in its vicinity. Typical X-ray properties of high-mass main-sequence stars ( $M_\star \geq 8~M_\odot$), where the emission originates from internal shocks in the radiation-driven stellar winds, are $L_{\rm X}/L_{\rm bol} \sim 10^{-7}$, soft X-ray spectra with typical temperatures of $kT \la 0.5$ keV and very little variability (Berghöfer et al. 1997). The X-ray properties we find for X1 to X7 are clearly different, they show hard X-ray emission ($kT \geq 3$ keV) and most of them also near-infrared excess, very similar to those observed for extremely young embedded objects (Class I protostars), which have typical plasma temperatures of $kT \sim 5{-}10$ keV (Feigelson & Montmerle 1999; Imanishi et al. 2001). Combining infrared data with pre-main-sequence evolutionary tracks (Palla & Stahler 1999), it is possible to estimate the approximate masses of some of the hard X-ray sources. Those estimates indicate that they are in the intermediate-mass regime of Herbig Ae/Be objects. Taking additionally into account the harder X-ray spectra compared with other Herbig Ae/Be studies, it is likely that the X-ray sources in IRAS 19410+2336 are even precursors of Herbig Ae/Be stars. The emitted X-ray photons with energies mostly above 2 keV indicate plasma temperatures >10⁷ K and X-ray luminosities around a few times 10³¹ erg s^-1. The latter values are well within the regime of Class I low-mass protostars (Feigelson & Montmerle 1999), but they are also consistent with the results obtained for Herbig Ae/Be stars (Zinnecker & Preibisch 1994). Thus, some of the objects are probably very young intermediate-mass pre-main-sequence sources, whereas other sources could also be low-mass Class I or T Tauri stars. The emission of one of the sources is consistent with X-ray variability.

In spite of the observation of hard X-ray emission in the weak-lined T Tauri star V773 Tau (Tsuboi et al. 1998), where the disk has already been dissipated to a large degree, it is unlikely that the hard X-ray spectra observed in younger class I sources are due to enhanced solar-type magnetic activity. Therefore, it is proposed that the hard X-ray emission, which is more often observed in class I sources than in weak-lined T Tauri stars, is produced by magnetic reconnection effects between the protostars and their accretion disks (Hayashi et al. 1996; Feigelson & Montmerle 1999; Montmerle et al. 2000). As the X-ray spectra of the intermediate-mass objects in IRAS 19410+2336 exhibit very similar signatures to such low-mass sources, our results are consistent with disks being present in intermediate-mass star formation as well.

For a better understanding of the nature of the underlying X-ray powering sources much work has to be done in the future. Deeper X-ray and near-infrared images will help to set stronger constraints on the physical properties of the sources: it will be necessary to obtain sensitive X-ray spectra to determine better the absorbing $N_{\rm {H}}$column densities and plasma parameters. It is also of great interest to further investigate the properties of the central and deepest embedded object, which means lowering the detection limits. Furthermore, the variability of the X-ray sources in very young massive star-forming regions is not known so far. Therefore, several approaches should be followed in the years coming: deep Chandra observations of the source of interest will disclose variabilities and faint emission of the central object. Additionally, a sample of similar sources has to be identified, because only a statistical analysis of several young high-mass star-forming regions can build a solid picture of the relevant physical processes. As high spatial resolution is essential for many of this studies, Chandra is a very promising choice. But considering the higher sensitivity of XMM-Newton, it might be possible to study grating X-ray spectra of the brightest sources of the sample of clusters studied then. On the near-infrared side, we suggest to get deeper images in the J, H and Kbands to improve the mass estimates of the X-ray emitting sources, and near-infrared spectroscopy might help classifying the types of stars (Hanson et al. 2002).

To summarize, X-ray studies of young massive star-forming regions are just in its infancy, and the next years with the space telescopes Chandra and XMM-Newton will bring many new insights in that research area. We also like to stress that multi-frequency studies over a wide range of bands are extremely promising approaches for the understanding of the physical processes forming massive stars.

Acknowledgements

We thank an anonymous referee for very helpful and detailed comments on the paper. This publication makes use of data products from the Two Micron All Sky Survey, which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center, funded by the National Aeronautics and Space Administration and the National Science Foundation. We also used data from the Digitized Sky Survey as provided by the Space Telescope Science Institute.