1 Introduction

In the past, star formation research in the X-ray regime has focused strongly on low-mass objects, e.g., T Tauri stars. These objects emit mainly in the soft range of the X-ray spectrum (<2 keV) with typical X-ray luminosities between 10²⁸-10³⁰ erg s^-1. The satellite observatory ROSAT was an ideal instrument to study such stars within a few 100 pc distance from the Sun, and the observed emission can be explained by enhanced solar-type magnetic activity (Feigelson & Montmerle 1999). In the last few years, a rising number of Class I protostars, which are still deeply embedded within their natal molecular cores ( $A_{\rm {V}} \approx 10$-100 mag), have been detected in the hard X-ray regime between 2 and 10 keV with X-ray luminosities higher than 10³⁰ erg s^-1. These detections were mostly made with the X-ray satellites ASCA and, most recently, with Chandra. Magnetic star-disk interactions are thought to be the most likely explanation for the hard X-ray emission (Hayashi et al. 1996; Montmerle et al. 2000). Furthermore, X-ray variability is observed in all types of low-mass pre-main-sequence objects. For an excellent recent review on these topics see Feigelson & Montmerle (1999). Recently, Tsuboi et al. (2001) and Tsujimoto et al. (2002) reported the first tentative detections of deeply embedded Class 0 protostellar candidates in OMC3 by Chandra.

In comparison with the low-mass regime, X-ray observations of massive star-forming regions have been rare. Due to the high visual extinction within such regions ( $A_{\rm {v}}$ up to a few 100 or even 1000), soft X-ray emission is completely absorbed by the gas along the line of sight. Hofner & Churchwell (1997) detected with ASCA for the first time hard X-ray emission in the massive star-forming region W3. Because of the low angular resolution of ASCA (>1') they could not determine whether the emission is caused by the superposition of many point sources, e.g., protostellar clusters, or whether it is due to a wind-shocked cavity resulting from strong stellar winds interacting with the surrounding medium. Recently, Churchwell (2001) reported that Chandra data of the same region with a spatial resolution of 0.5'' resolve the emission into many individual sources distributed over the entire W3 complex. Similarly, Garmire et al. (2000) and Feigelson et al. (2002) reported about 1000 X-ray emitting pre-main-sequence stars between 0.05 and $50~M_{\odot}$in the Orion Nebula. Additionally, Zinnecker & Preibisch (1994) found X-ray emission with the ROSAT satellite in the soft X-ray band associated with several intermediate-mass Herbig Ae/Be pre-main-sequence stars. The derived X-ray luminosities for the Herbig Ae/Be stars are ranging between 10³⁰ erg s^-1 and 10³² erg s^-1. Preibisch & Zinnecker (1995) speculated that the emission might originate from coronal activity due to shear dynamo action. A study of the more distant molecular clouds Monoceros and Rosette also found indirect evidence for X-ray emission from intermediate-mass pre-main-sequence sources (Gregorio-Hetem et al. 1998). An example of hard X-ray emission from a Herbig Be star is MWC 297 (Hamaguchi et al. 2000). In a recent X-ray study of the Monoceros R2 molecular cloud, Kohno et al. (2002) detected X-ray emission for stars of all masses, in particular hard X-ray emission from high-mass pre-main-sequence or Zero-age-main-sequence stars. Based on these results, X-ray emission seems likely to be an ubiquitous phenomenon in the protostellar evolution of stars of all masses.

Figure 1: Left: the contours present the large-scale 1.2 mm dust continuum emission from $10\%$ to $90\%$ (steps of $10\%$) of the peak flux (Beuther et al. 2002a). The grey-scale shows the K-band image. Right: the contours show the PdBI 2.6 mm continuum dust cores (black: levels 3.6( $1.2=1\sigma$)8.4 mJy/beam, white: 9.6(2.4)20 mJy/beam) superposed on the K-band image. Circles and stars mark the X-ray sources presented in this paper. Sources with circles are most likely associated with the massive star-forming region, whereas the asterisks represent those that might be foreground objects. The box outlines the region presented in Fig. 4.

So far, X-ray studies in high-mass star-forming regions focused on more evolved massive star formation sites, while the very deeply embedded phase - and thus the youngest stage of stellar evolution - has not been detected at all. Results obtained in recent years by our group (Sridharan et al. 2002; Beuther et al. 2002a,b,d) and other groups studying massive star formation (e.g., Cesaroni et al. 1997; Zhang et al. 2002; Tan & McKee 2002; Yorke 2002) support the hypothesis that massive stars form via disk accretion in a similar fashion as low-mass stars. Therefore, high-mass star-forming cores are promising candidate regions where in rather small spatial areas a number of sources could be hard X-ray emitters via the physical process of star-disk interactions (Hayashi et al. 1996; Montmerle et al. 2000). Necessary observational requirements are first of all sensitivity in the hard X-ray regime (>2 keV), because only hard X-ray photons can penetrate high gas column densities. Additionally, high angular resolution is needed to resolve different sub-sources of the forming cluster. While ROSAT was not sensitive to hard X-ray photons, the spatial resolution of ASCA was not sufficient to study massive star-forming regions in detail at their typical distances of a few kpc. The new-generation X-ray satellite telescopes Chandra and XMM-Newton comprise both features, being sensitive up to 10 keV, and having a spatial resolution of 0.5'' (Chandra) and 15''(XMM-Newton), respectively. Especially Chandra is able to resolve many different sub-sources as impressively demonstrated in Orion by Garmire et al. (2000) and Feigelson et al. (2002).

Here, we present a Chandra X-ray study of the very young, massive and deeply embedded star-forming cluster IRAS 19410+2336. IRAS 19410+2336 is part of a large sample of 69 high-mass protostellar candidates which has been studied extensively in a series of papers during the last years (Sridharan et al. 2002; Beuther et al. 2002a-d). We assume the source to be located at its near kinematic distance of 2.1 kpc (Sridharan et al. 2002), because the derived outflow parameters are unreasonably high for the far kinematic distance (Beuther et al. 2002b). At the distance of 2.1 kpc, its infrared derived luminosity is $10^4~L_\odot$, and one observes two adjacent star-forming cores with masses of $840~M_{\odot}$ and $190~M_{\odot}$(Sridharan et al. 2002; Beuther et al. 2002a). Each core drives a massive bipolar outflow in east-west direction (Beuther et al. 2002b). At the center of the southern massive core, a very compact and weak ($\sim$1 mJy) cm wavelength source is detected, which coincides with H₂O and Class  II CH₃OH maser emission (Beuther et al. 2002c). Figure 1 (left) gives an overview of the region of interest with the 1.2 mm dust continuum data (Beuther et al. 2002a) superposed on an infrared K-band image (Sect. 2.3). As the source is located in the Galactic plane, confusion due to foreground and background sources is expected and has to be disentangled by the different observations. We focus on the X-ray emission of this region and correlate the detected X-ray sources with high-resolution images in the mm and near-infrared regime. Section 2 describes the different observations we performed (X-ray, near-infrared and mm data), and in Sect. 3 we derive the main physical parameters of this star-forming region from our data (source detections, spectra, X-ray luminosities, plasma temperatures and masses). Finally, Sect. 4 compiles our conclusions and puts the observational findings into a more general framework of high-mass star formation.

 

 

Table 1: Compilation of the parameters derived from the X-ray and 2MASS data: positions of the X-ray sources, number of counts per source, mean energy of detected photons, best fits of Raymond-Smith plasma temperatures (kT [keV] and T [K]) and derived X-ray luminosities $L_{\rm {X}}$ with error ranges. Colons mark less conclusive fits. Additionally, the column densities $N_{\rm {H}}$ (and error ranges) from the X-ray fits and the 2MASS derived extinctions $A_{\rm {v}}$ are presented.

source	RA	Dec	#	mean	kT	T	$L_{\rm {X}}$	$N_{\rm {H}}$	$A_{\rm {v}}$
	(J2000.0)	(J2000.0)	[cts]	[keV]	[keV]	[107 K]	[1031 erg s-1]	[1022 cm-2]
X1	19:43:10.8	+23:44:04.6	17	2.6	5.6 [>4 ]	6.5 [>4.6 ]	1.1 [0.7-1.4]	1.0 [0.7-1.3]	17:
X2	19:43:11.0	+23:44:16.1	10	3.9	9.6 [>6 ]	11.1 [>7 ]	2.1 [1.5-2.8]	5.9 [4.5-8.0]	-
X3	19:43:09.6	+23:43:57.7	19	3.5	10: [>6 ]	11.6:[>7 ]	2.4 [1.9-3.0]	2.9 [2.3-3.8]	-
X4	19:43:09.4	+23:44:01.0	17	2.5	3.2 [2.2-5.3]	3.7 [2.6-6.2]	1.2 [0.8-1.7]	1.1 [0.9-1.4]	-
X5	19:43:12.5	+23:44:21.8	9	5.3	9.3 [>5 ]	10.8 [>5.8 ]	2.6 [1.6-3.5]	10: [>7.2 ]	-
X6	19:43:11.8	+23:43:53.0	12	3.8	7.4 [>5 ]	8.6 [>5.8 ]	1.8 [1.2-2.3]	3.7 [2.6-5.3]	18
X7	19:43:11.5	+23:43:36.5	7	3.6	7.5 [>4 ]	8.7 [>4.6 ]	0.8 [0.5-1.1]	2.7 [1.8-3.9]	12
X8	19:43:10.6	+23:44:56.8	12	2.2	1.4 [1.1-1.9]	1.6 [1.3-2.2]	2.2 [2.0-2.8]	3.0 [2.5-3.7]	17
X9	19:43:08.6	+23:44:24.7	9	2.8	1.2:	1.4:	-	0.3:	12
X10	19:43:08.9	+23:44:36.6	19	2.2	1.5:	1.7:	-	0.8:	3
X11	19:43:07.7	+23:44:49.3	7	1.2	1.0:	1.2:	-	0.2:	-
X12	19:43:14.3	+23:45:14.9	12	1.7	1.3 [1.1-1.7]	1.5 [1.3-2.0]	-	0.6 [0.4-0.8]	1
X13	19:43:13.4	+23:45:15.9	13	1.2	1.3 [1.1-1.5]	1.5 [1.3-1.7]	-	0.3 [0.2-0.4]	2