Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.arcetri.astro.it/science/starform/preprints/cesa_23.pdf Äàòà èçìåíåíèÿ: Fri Nov 16 14:36:41 2012 Äàòà èíäåêñèðîâàíèÿ: Mon Feb 4 13:49:57 2013 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: http astrokuban.info astrokuban

November 16, 2012

A close-up view of a bipolar jet: Sub-arcsecond near-IR imaging of the high-mass protostar IRAS 20126+4104
R. Cesaroni1 , F. Massi1 , C. Arcidiacono1,2 , M.T. Beltran1 , D. McCarthy3 , C. Kulesa3 , K. Boutsia4 , D. Paris4 , F. ´ Quiros-Pacheco1 , and M. Xompero1 ´
1

2 3 4

INAF, Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, I-50125 Firenze, Italy e-mail: cesa@arcetri.astro.it, fmassi@arcetri.astro.it, mbeltran@arcetri.astro.it, fquiros@arcetri.astro.it, marco@arcetri.astro.it INAF, Osservatorio Astronomico di Bologna, Via Ranzani 1, I-40127 Bologna, Italy e-mail: carmelo.arcidiacono@oabo.inaf.it Steward Observatory, The University of Arizona, 933 N. Cherry Ave., Tucson, AZ-85721, U.S.A. e-mail: dwmccarthy@gmail.com, ckulesa@as.arizona.edu INAF, Osservatorio Astronomico di Roma, via Frascati 33, I-00040, Monteporzio, Italy e-mail: konstantina.boutsia@oa-roma.inaf.it, diego.paris@oa-roma.inaf.it

Received date; accepted date
ABSTRACT

Context. The formation of OB-type stars up to (at least) 140 M can be explained via disk-mediated accretion and in fact growing observational evidence of disk-jet systems is found in high-mass star-forming regions. Aims. With the present observations we wish to investigate at sub-arcsecond resolution the jet structure close to the well studied high-mass protostar IRAS 20126+4104, which is known to be surrounded by a Keplerian disk. Methods. Adaptive optics imaging of the 2.2 µm continuum and H2 and Br line emission have been performed with the Large Binocular Telescope, attaining an angular resolution of 90 mas and an astrometric precision of 100 mas. Results. While our results are consistent with previous K-band images by other authors, the improved (by a factor 3) resolution allows us to identify a number of previously unseen features, such as bow shocks spread all over the jet structure. Also, we confirm the existence of a bipolar nebulosity within 1 from the protostar, prove that the emission from the brightest, SE lobe is mostly due to the H2 line, and resolve its structure. Conclusions. Comparison with other tracers such as masers, thermal molecular line emission, and free-free continuum emission proves that the bipolar nebulosity is indeed tracing the root of the bipolar jet powered by the deeply embedded protostar at the center of the Keplerian disk.
Key words. Stars: formation ­ ISM: individual objects: IRAS 20126+4104 ­ ISM: jets and outflows

1. Introduction
In the context of star formation, high-mass stars are commonly defined as those exceeding 8 M . While the value of such a threshold may depend on a number of physical parameters, its existence is determined by the simple fact that the radiation pressure of luminous stars may halt the accretion process, thus questioning the possibility that high-mass stars can form through (disk) accretion like their low-mass siblings. However, recent observational and theoretical studies indicate that diskjet systems may be a viable solution to explain the formation of OB-type stars. On the one hand, growing evidence of circumSend offprint requests to: R. Cesaroni, e-mail: cesa@arcetri.astro.it Based on observations carried out with the Large Binocular Telescope. The LBT is an international collaboration among institutions in the United States, Italy and Germany. LBT Corporation partners are: The University of Arizona on behalf of the Arizona university system; Istituto Nazionale di Astrofisica, Italy; LBT Beteiligungsgesellschaft, Germany, representing the Max-Planck Society, the Astrophysical Institute Potsdam, and Heidelberg University; The Ohio State University, and The Research Corporation, on behalf of The University of Notre Dame, University of Minnesota, and University of Virginia.

stellar (Keplerian) disks with associated bipolar jets along the rotation axis is being obtained by means of high angular resolution imaging at (sub)mm and IR wavelengths (see e.g. Cesaroni et al. 2007, Kraus et al. 2010, de Wit et al. 2010, 2011). On the other hand, numerical models confirm that disk-mediated accretion can explain the formation of stars as massive as 140 M (Krumholz et al. 2009; Kuiper et al. 2010, 2011). The next step of this study will be the analysis of the accretion (through the disk) and ejection (through the associated jet) processes close to the (proto)star. While interferometric observations at (sub)mm wavelengths have been traditionally used to investigate the heavily extincted surroundings of deeply embedded OB (proto)stars, substantial progress has been recently made in the near-IR regime. In particular, adaptive optics provides us with angular resolutions <100 mas in the near-IR, su perior to those attainable until now in the (sub)mm. Although at short wavelengths one does not expect to detect the (proto)star at the center of the disk, if the line of sight is favourable it should be possible to see the scattered stellar photons from the innermost part of the disk/jet and trace the jet itself through the H2 line emission. With this in mind, we have performed LBT/PISCES AO observations of the 2.2 µm continuum and H2 v =1­0 S(1) line


N E
1arcsec

IRAS20126+4104
LBT FLAO+PISCES Brg+Ks+H2

Fig. 1. Composite colour image obtained by combining the emission from the three filters used: H2 filter (red), Ks (green), and Br (blue). Regions dominated by continuum emission appear white in the image, while those traced by the H2 line are red. Br emission is not detected.

emission towards the high-mass protostar IRAS 20126+4104. This is a well studied object of 104 L , associated with a diskjet/outflow system and located at a distance of 1.64±0.05 kpc (Moscadelli et al. 2011). The disk of 1000 AU is undergoing Keplerian rotation consistent with a stellar mass of 7 M (Cesaroni et al. 2005). The jet/outflow has been imaged from the small (a few 100 AU) to the large (0.5 pc) scale using a variety of tracers (H2 O, H2 , free-free continuum, SiO, HCO+ , CO, ´ etc.; Cesaroni et al. 1997, 1999; Lebron et al. 2009; Moscadelli et al. 2000, 2005, 2011; Su et al. 2007; Caratti o Garatti et al. 2008; Hofner et al. 1999, 2007) and is found to undergo precession about the rotation axis of the disk (Shepherd et al. 2000; Cesaroni et al. 2005). Near-IR images in the K, L, and M' bands with 0. 3 resolution were obtained by Sridharan et al. (2005; hereafter SWF), unveiling the existence of a bipolar nebulosity along the direction of the jet/outflow. At K band this bipolar structure outlines a dark lane which is interpreted by the authors as the disk silouhette, whereas in the other bands the authors claim to detect emission from the disk as well as from a nearby companion. However, later observations performed by De Buizer (2007) with an 8-m class telescope in the mid-IR have not confirmed the presence of emission from the disk and questioned the presence of the secondary source. In contrast, more recently, de Wit et al. (2009) imaged the source with Subaru at 24.5 µm, revealing strong, compact emission that could originate from the dark lane. The scope of our new observations is to significantly improve (by a factor 3) on the angular resolution of previous IR images and thus shed light on the nature of the structures observed within a few 1000 AU from the IRAS 20126+4104 protostar.

12 west of the main source, as the reference for the AO loop, closed using 66 modes. The on-source frames were first flat-fielded, sky subtracted with a median of the sky frames and corrected for the optical PISCES distortion using the "drizzle" IRAF routine. Finally, they were registered and combined together. The final images were calibrated using the only two stars from the 2MASS catalogue falling inside the field. In principle, this makes the calibration prone to stellar variability, but we checked that the measured flux from the jet in the H2 line is consistent within 30% with the values reported by Cesaroni et al. (1997). To maximize the astrometric precision, we retrieved a much larger K image of the field from NICS at the TNG (better resolved than 2MASS) and used a number of stars within 2 of the jet, to obtain a positional fit to the corresponding 2MASS stars. We used the astrometric solution (through IRAF routines) to refine the absolute positions of the pair of 2MASS stars inside the PISCES field, yielding a correction of only a fraction of arcsec to the coordinates given in the 2MASS Catalogue. The astrometric precision of the final images should be better than 0. 1 and the point spread function is 0. 09. The Br image does not display line emission, so it was used only to estimate the continuum contribution inside the H2 filter and will not be mentioned anymore in the following sections. A continuum-free H2 image was constructed by subtracting the Br image after shifting and scaling.

3. Results and discussion
3.1. The large scale structure of the jet

2. Observations and data reduction
The data were collected on June 21, 2012, using the PISCES Near Infrared Camera (McCarthy et al. 2001) installed at the focal plane of the First Light Adaptive Optics system (Esposito et al. 2010, 2011) of the Large Binocular Telescope. This allows a field of 21 â 21 to be imaged with a plate scale of 0.02 arcsec/pixel. The observations were carried out through the Ks broad-band filter (at 2.12 µm), and the H2 (2.12µm) and Br (2.17 µm) narrow-band filters. We used a star with R 14 mag ((J2000)=20h14m 25s.069, (J2000)=+4113 32. 23),

Our results are illustrated in Figs. 1 and 2. The former is a composite image where the three filters have been combined. The latter shows the continuum-subtracted H2 line and the broad band (Ks ) images separately. Even by-eye comparison between the two panels of this figure clearly reveals that the Ks image contains substantial continuum emission. This outlines a bipolar nebulosity presumably caused by scattered stellar photons along the outflow axis, in agreement with the findings of Cesaroni et al. (1997) ­ see their Fig. 12. More interesting is the image of the H2 line emission (Fig. 2b). Here one sees that the whole emitting region is spotted with bow shocks, spanning a broad range of sizes. This fact as well as the highly irregular structure of the jet lobes suggests that the ejection phenomenon is episodic, irregular in time, and asymmetric with respect to the star. Worth of notice is also the wiggling shape of the lobes, consistent with the jet being precessing. As noted by Caratti o Garatti et al. (2008), the resulting pattern is too complex to be fitted with a simple precession law and suggests the presence of multiple stars perturbing the trajectory of the jet/outflow axis. The material launched in this way is bound to clear up a relatively broad solid angle along the jet/outflow axis, thus allowing the stellar photons to escape from the star neighbourhoods. This can explain why in Fig. 2a the continuum emission due to scattered stellar photons is seen at large distances from the star. An intriguing feature worthy of mention, is the "mushroomshaped" structure located a few arcsec to the NW from the jet center. This is shown in Fig. 3, where we plot the free-free continuum map by Hofner et al. (2007) and the water maser spots detected by Moscadelli et al. (2005, 2011) with the associated proper motions. Also shown is the nominal position of the protostar, obtained from a model fit to the H2 O masers (Moscadelli et al. 2011). The latter lies along the symmetry axis (dashed line in the figure) of the "mushroom-shaped" feature, within the as-


Fig. 2. a. Image of the emission in the Ks filter of IRAS 20126+4104. The white rectangle outlines the region displayed in Fig. 5. b. Same as above, for the H2 line emission. The dotted circles indicate artifacts due to residual emission after the continuum subtraction. The arrow points at the "mushroom-shaped" structure shown in Fig. 3.

trometric uncertainty of the LBT image (<0. 1). The shape and orientation of this feature, compared to the other jet tracers, especially the H2 O masers, strongly suggest that the H2 emission belongs to the jet and outlines a bow shock. It is worth noting the impressive similarity between the "mushroom-shaped" feature and the simulation of a magnetized jet powered by a disk wind, by Staff et al. (2010; see e.g. their Figs. 1 and 2, displaying gas density and [OI] line intensity maps). Although the latter refers to scales of only 60 AU from a low-mass star, one cannot rule out the possibility that more massive stars can produce analogous structures at much larger distances (in our case 4000 AU) and with larger sizes.
3.2. Nature of the small-scale bipolar feature

Fig. 3. Image of the H2 line emission from the IRAS 20126+4104 region, with overlayed the 3.6 cm continuum map (dotted contours) from Hofner et al. (2007) and the H2 O maser spots (green points) with associated proper motions (arrows) from Moscadelli et al.(2011). The solid contours correspond to the intensity levels marked in the colour scale on top of the figure. The yellow star marks the nominal position of the protostar, obtained from the model fit of Moscadelli et al. (2011) to the H2 O maser positions and velocities. The dashed line is the symmetry axis of the "mushroom-shaped" H2 line emitting feature in the top right of the figure.

In the following, we focus our attention on the K-band bipolar structure that SWF assert to be tracing the jet very close to the circumstellar disk, seen as a dark silouhette across the two jet lobes. To check the consistency between our data and those of SWF, in Fig. 4 we present an image of the Ks band emission that covers the same region displayed in the right panel of Fig. 1 of SWF, smoothed to the same resolution (0. 3). The similarity between the two images is quite satisfactory, although the NW lobe appears slightly shifted to the south in our data: this is not surprising, as this lobe is only barely detected in the non smoothed image (see Fig. 5a). Besides being very faint in the 2.2 µm continuum, this lobe is not seen in the H2 line emission (see Fig. 5b), which might cast some doubt on its nature. However, we believe that also the NW lobe is associated with a bow-shock cavity, as well as the SE lobe, but the H2 line is not detected because of extinction. This hypothesis is supported by SWF, who estimate a larger extinction for the NW lobe (see their Table 1); moreover, the mid-IR images of De Buizer (2007) demonstrate that the NW lobe becomes progressively more prominent at longer wavelengths, where the dust opacity decreases. This result is

confirmed by de Wit et al. (2009), who could image the NW lobe at 24.5 µm, but failed to reveal the SE lobe. Our observations are a substantial improvement with respect to those of SWF, not only because of better angular resolution, but also because we can discriminate how much of the K-band emission detected by SWF is due to the H2 line. Comparison between Fig. 5a and 5b shows that the weak NW nebulosity is basically made of continuum emission, whereas the SE nebulosity is dominated by the line emission, with only a faint continuum contribution in the south-western part of it. With the aim to shed light on the nature of this bipolar structure, we compare our images with other disk and jet/outflow tracers from previous studies of IRAS 20126+4104. This is done in Fig. 6, where we overlay the free-free 3.6 cm continuum map of Hofner et al. (2007) and the H2 O and CH3 OH spots detected by various authors, on our H2 image. We also plot the peaks of the CH3 CN(12­11) and C34 S(5­4) emission that according to Cesaroni et al. (1999, 2005) trace a Keplerian circumstellar disk. These peaks have been obtained by fitting the line emission at different velocities with a 2-D Gaussian (see e.g. Fig. 5 of Cesaroni et al. 1999). Since the astrometric uncertainty of the interferometric observations of CH3 CN and C34 S is significantly worse (0. 4) than that of the VLBI and VLA observations of the masers and 3.6 cm continuum, we have decided to recenter the CH3 CN and C34 S in the following way. Besides the molecular line emission, an image of the mm continuum emission was obtained in the interferometric observations by Cesaroni et al. (1999, 2005). We made the reasonable assumption that this coincides with the position of the protostar and applied a shift to each continuum map in such a way that the


Fig. 4. Image of the mean emission in the Ks band smoothed to the same angular resolution (0. 3) as the data of SWF. The region displayed corresponds to that covered in the left panel of Fig. 1 of SWF. The angular resolution is indicated by the circle in the bottom right of the figure.

Fig. 5. Enlargment of the central region of Figs. 2a and 2b.

as proved by their proper motions (see Fig. 3 of Moscadelli et al. 2011). Note also that the coincidence between the CH3 CN and C34 S peaks with (part of) the CH3 OH spots lends support to our method to improve on the astrometry of the mm data. An interesting result depicted in Fig. 6 is the striking asymmetry between the SE and NW regions with respect to the protostar. While there is little doubt that the masers and free-free emission are tracing the NW lobe of the jet (apart from an H2 O spot with proper motion directed towards SE, likely associated with the SE lobe), the origin of the IR emission is more doubtful. This is only barely detected to the NW, but very strong to the SE, which seems inconsistent with the orientation of the jet, because the emission from the NW lobe of the jet is blue shifted and hence pointing towards the observer (Moscadelli et al. 2011; Cesaroni et al 2005). If the IR emission is associated with the jet, it should be more easily detected from the lobe pointing towards us (the NW lobe), than from that pointing away from us (the SE lobe), because the latter is obviously more extincted. How can one reconcile this scenario with the strong H2 line emission seen to the SE and the lack of emission to the NW? First of all, the inclination of the jet axis with respect to the plane of the sky is very small, about 10 according to Moscadelli et al. (2011). Consequently, the difference in extinction between the SE (red shifted) and NW (blue shifted) lobes cannot be justified in terms of different orientation with respect to the observer (the NW lobe pointing towards us, the other away from us). Therefore, the (small) jet inclination cannot explain why the H2 line (and, to some extent, continuum) emission is much brighter on one side than on the other. In our opinion, the explanation is related to the nature of ejection phenomenon. In Sect. 3.1, based on the structure of the H2 emission on a large scale, we have concluded that the jet is asymmetric, irregular, and precessing. This implies that the ejection phenomenon not only is discontinuous and non-periodic, but also occurs at different times on the two sides with respect to the protostar. Moreover, due to precession, every new ejection has to dig its way through new, fresh circumstellar material. If the NW lobe corresponds to a more recent ejection episode than the SE lobe, it is not surprising that the jet is still pearcing a narrow passage throught the dense molecular neighbourhoods. Vice versa, on the SE side an older ejection episode may have had enough time to blow the dense gas away, creating an inflated cavity from which the IR photons can escape. The absence of dense gas explains why no thermal jet and almost no maser emission is seen to the SE. We conclude that the nature of the near-IR emission within 1 from the protostar is coming from the bipolar jet directed SE­ NW, in agreement with the interpretation of SWF. Our results demonstrate the potential of AO observations in the near-IR even in regions characterized by large local extinction and call for observations at similar wavelengths (e.g. in the L and M' bands; Sridharan et al. 2007) to further the investigation of the small scale structure and stellar content of this interesting object.
Acknowledgements. It is a pleasure to thank Leonardo Testi for providing us with the TNG images used to improve the astrometric accuracy of our data. We are also grateful to the referee, Willem-Jan de Wit, for his prompt reply and constructive criticisms. We acknowledge the support from the LBT-Italian Coordination Facility for the execution of observations, data distribution and reduction.

continuum peak coincides with the the nominal stellar position ­ obtained from the model fit to the H2 O masers (Moscadelli et al. 2011). The same shift was applied to the corresponding line peaks, as the mm line and continuum data were acquired simultaneously and the relative positional error is negligible. Inspection of Fig. 6 confirms the scenario depicted by SWF, as the disk lies right at the border of the IR nebulosity outlined by both H2 and (weak) continuum emission. Interestingly, the CH3 OH masers appear to be partly lying on the surface of the disk and partly lifting from it, parallel to the direction of the jet,

References
Caratti o Garatti, A., Froebrich, D., Eisloffel, J., Giannini, T., & Nisini, B. 2008, ¨ A&A, 485, 137 Cesaroni, R., Felli, M., Testi, L., Walmsley, C.M., & Olmi, L. 1997, A&A, 325, 725


Fig. 6. Image of the mean emission in the H2 filter (line plus continuum) of the central region of IRAS 20126+4104, with overlayed the 3.6 cm continuum map (dashed contours) from Hofner et al. (2007), the CH3 OH maser spots (cyan points) from Moscadelli et al. (2011), and the H2 O maser spots (green points) from Moscadelli et al. (2005, 2011), Edris et al. (2005), and Trinidad et al. (2005). The white points denote the peaks of the CH3 CN(12-11) and C34 S(5­4) line emission from Cesaroni et al. (1999, 2005), whose position has been realigned as explained in the text. The green arrows indicate the direction and speed of the H2 O maser proper motions. The yellow star marks the nominal position of the protostar, obtained from the model fit of Moscadelli et al. (2011) to the H2 O masers.
Cesaroni, R., Felli, M., Jenness, T., et al. 1999, A&A, 345, 949 Cesaroni, R., Neri, R., Olmi, L., et al. 2005, A&A, 434, 1039 Cesaroni, R., Galli, D., Lodato, G., Walmsley, C.M., & Zhang, Q. 2007, in Protostars and Planets V, ed. B. Reipurth, D. Jewitt, & K. Keil (Tucson: Univ. of Arizona Press), 197 De Buizer, J.M. 2007, ApJ, 654, L147 de Wit, W.J., Hoare, M.G., Fujiyoshi, T., et al. 2009, A&A, 494, 157 de Wit, W.J., Hoare, M.G., Oudmaijer, R.D., & Lumsden, S.L. 2010, A&A, 515, A45 de Wit, W.J., Hoare, M.G., Oudmaijer, R.D., et al. 2011, A&A, 526, L5 Edris, K.A., Fuller, G.A., Cohen, R.J., Etoka, S. 2005, A&A, 434, 213 Esposito, S., Riccardi, A., Quiros-Pacheco, F., et al. 2010, ApOpt, 49, G174 ´ Esposito, S., Riccardi, A., Pinna, E., et al. 2011, in Astronomical Adaptive Optics Systems and Applications IV, ed. R.K. Tyson, M. Hart, Proceedings of the SPIE, 8149, 814902 Hofner, P., Cesaroni, R., Rodr´ iguez, L.F., & Mart´ J. 1999, A&A, 345, L43 i, Hofner, P., Cesaroni, R., Olmi, L., et al. 2007, A&A, 465, 197 Kraus, S., Hofmann, K.-H., Menten, & Karl M. 2010, Nature, 466, 339 Krumholz, M.R., Klein, R.I., McKee, C.F., Offner, S.S.R., & Cunningham, A.J. 2009, Science, 323, 754 Kuiper, R., Klahr, H., Beuther, H., & Henning, Th. 2010, ApJ, 722, 1556 Kuiper, R., Klahr, H., Beuther, H., & Henning, Th. 2011, ApJ, 732, 20 Lebron, M., Beuther, H., Schilke, P., & Stanke, Th. 2006, A&A, 448, 1037 ´ McCarthy, D.W., Jr., Ge, J., Hinz, J.L., Finn, R.A., & de Jong, R.S. 2001, PASP, 113, 353 Moscadelli, L., Cesaroni, R., & Rioja, M.J. 2000, A&A, 360, 663 Moscadelli, L., Cesaroni, R., & Rioja, M.J. 2005, A&A, 438, 889 Moscadelli, L., Cesaroni, R., Rioja, M.J., Dodson, R., & Reid, M.J. 2011, A&A, 526, A66 Shepherd, D.S., Yu, K.C., Bally, J., Testi, L. 2000, ApJ, 535, 833 Sridharan, T.K., Williams, S.J., & Fuller, G.A. 2005, ApJ, 631, L73 (SWF) Sridharan, T.K., Saito, M., & Fuller, G.A., & Kandori, R. 2007, Bulletin of the American Astronomical Society, 39, 849 Staff, J.E., Niebergal, B.P., Ouyed, R., Pudritz, R.E., & Cai, K. 2010, ApJ, 722, 1325 Su, Y.-N., Liu, S.-Y., Chen, H.-R., Zhang, Q., & Cesaroni, R. 2007, ApJ, 671, 571 Trinidad, M.A., Curiel, S., Migenes, V., et al. 2005, AJ, 130, 2206