RESEARCH ARTICLES

IRCAM3 Imaging of Outflows from Young Stars

Chris Davis, Tom Ray, David Corcoran

Dublin Institute for Advanced Studies, EIRE

Jochen Eisloffel

Thuringer Landessternwarte, Tautenburg, Germany

Energetic winds from T-Tauri stars and more embedded infrared/radio sources are a phenomenon inherent to the star-formation process. By reversing the process of infall, they provide a mechanism by which a collapsing and rotating cloud envelope may shed angular momentum, they provide pressure support for the parent cloud and ultimately they may determine the final mass of the star.

Mass outflows from regions of star formation are observable via molecular tracers, typically CO rotational lines, and via the optical and infrared emission lines produced in shocks within each flow and as the flow interacts with its molecular surroundings. We know these shocks as the emission-line nebulae Herbig-Haro Objects (HH Objects), though many HHs are now recognised as simply the brightest knots within more continuous high-velocity, highly collimated stellar jets.

Yet the very youngest jet systems, those still deeply embedded within their natal cores, are only observed at infrared wavelengths. These jets are thought to be actively driving massive, molecular (CO) outflows. As a first step towards a better understanding of this process, we have conducted an imaging survey of a number of outflows. Here we report on some recent results from UKIRT.

Narrow-band images in the H₂ v=1-0 S(1) line (and in nearby continuum) were obtained with IRCAM3 in December 1994. The default pixel scale of 0.286 arcsec/pixel was used which meant that extensive mosaicing was needed to cover the outflow regions we were interested in.

Two of the most interesting regions observed were HH24-26 and L1634. The HH objects 24, 25 and 26 are situated a few arcminutes south of NGC2068 in L1630 (distance ~ 400 pc). HH24 is a well-known complex of at least three collimated optical outflows (Mundt, Ray & Raga 1991, ApJ, 319, 273), while HH25 and HH26 appear as more nebulous HH objects a few arcminutes to the south (Jones et al. 1987, AJ, 94, 1260). L1634, on the other hand, is a little-known molecular outflow (to date no CO maps have appeared in the literature). L1634 itself is a dark cloud in Orion (distance ~ 500 pc) which is perhaps best known as the home of the HH Object RNO40 (recently designated HH240).

HH24-26

The H₂ (+ continuum) image in Figure 1 of the HH24-26 region reveals a number of distinct outflows. Four of these are identified with known HH outflows (HH24A, HH24B, HH25 and HH26), though we also observe a cusp of H2 emission extending northwestward from SSV61 and a jet asociated with HH24-MMS, the youngest star in the region (and indeed one of the youngest stars known). The image shows a number of shock regions that are not evident in optical images (presumably because of extinction). The HH25 and HH26 outflows are particularly striking, being almost orthogonal to one another. It is tempting to suggest that the two flows collide near SSV60. This seems unlikely, however, since neither is diverted nor decollimated beyond the region where they apparently cross. The shock features in both HH25 and HH26 describe smooth arcs which may be due to ambient density gradients or large scale magnetic fields. Alternatively, they may reflect the proper motions of their sources.

FIGURE 1 : A narrow band IRCAM3 image, at 2.122 microns (H₂ + continuum), of the HH24-26 region. HH26-IR powers a spectacular east-west molecular outflow; the north-south HH25 outflow (driven by HH25-MMS) crosses this flow near SSV60. Note also the "H₂-jet" associated with HH24-MMS.

Two of the outflow sources in this region, HH24-MMS and HH25-MMS (marked with crosses), are more heavily embedded and presumably less evolved than their near-infrared neighbours. Indeed, HH24-MMS is believed to be a very young, ``Class 0'' source. We find that both sources power molecular outflows which are delineated by shocked, H<sub>2</sub> line emission. A weak, ``H₂ jet'' extends northeast-southwest from HH24-MMS while HH25-MMS drives a more extensive molecular outflow.

FIGURE 2 : H₂ (+ continuum) image of the HH25 and HH26 outflows with, superimposed, a contour plot showing the CO J=2-1 outflows mapped by Gibb & Heaton (1993). The full contours show the blue-shifted CO gas, the dashed contours the red-shifted gas.

The obvious association between the near-infrared HH knots in HH25 and HH26 and the outflows mapped at the JCMT in CO J=2-1 by Gibb & Heaton (1993, MNRAS, A&A, 276, 511) is quite striking (Figure 2). Indeed, the slightly confusing picture presented by the CO maps alone is clarified when a comparison is made with the near-infrared data. We see a perfect alignment between the string of HH knots in HH26 and its associated CO outflow. In an almost orthogonal direction, the knots in HH25 are similarly aligned with a CO outflow. On close inspection, we see that many of the peaks and bumps in the CO maps are associated with H₂ shock features. For example, the northerly extension of the blue-shifted lobe of the almost east-west HH26 CO outflow (near HH25-MMS) is almost certainly due to the overlapping HH25A outflow here, as is the bump on the southern side of the blue-shifted HH26 lobe. We are also left in little doubt that HH25-MMS and HH26-IR are the exciting sources of the outflows, since,

1. they lie on their outflow axes (as seen in H₂),
2. they are situated between their respective blue- and red-shifted CO outflow lobes.

In HH25 and HH26 the brightest H₂ peaks are generally coincident with or just downwind of the peaks in the CO outflow maps. This coincidence has recently been identified in a number of other outflows (Davis & Eisloffel 1995, A&A, 300, 851). Like Davis & Eisloffel, we interpret this as support for the ``prompt entrainment'' mechanism for jet-driven molecular outflows. This model predicts that massive molecular (CO) outflows may be driven by a collimated stellar jet, as gas is swept up in bow shocks formed at the head of or along the length of the jet. This mechanism will also produce clumps of molecular material just behind the shocks, as is generally observed here. That fact that the energy radiation rate in the observed molecular shocks is comparable to the mechanical power in the molecular (CO) outflow also lends support to this theory (see our forthcomming paper for a more detailed discussion; Davis et al. 1997, A&A, submitted).

L1634

Like the ``H₂-jet'' in HH24 and the HH25 outflow, the L1634 outflow is powered by a deeply embedded source that is not detected at near-IR wavelengths. This source is therefore probably very young (extreme Class I or Class 0). However, the near-IR image in Figure 3 does reveal many bright, extended features along the outflow axis. All of these are H₂ line emission features, the exception being the S-shaped wisp around the star situated some 40 arcsec east of the outflow source. The H₂ features to the east of the exciting source (HH241A-D) reside in the red-shifted lobe of the outflow; those to the west of the source (HH240A-D) are in the blue-shifted lobe.

FIGURE 3 : A 2.122 micron (H₂ + continuum) IRCAM3 image of the L1634 outflow. The near-IR HH objects HH240A-D and HH241A-D lie in the blue- and red-shifted lobes of the outflow respectively.

It is interesting to note that the eastern and western lobes of the outflow, as traced in H2, are equal in length. The distance on the sky from the source to HH240D and HH241D is 180 arcsec (0.40 pc) and 173 arcsec (0.39 pc) respectively. Also, for each of the bright H₂ structures in the western lobe (HH240A-D), there is an ``equivalent'' structure, at roughly the same distance from the outflow source, in the eastern lobe (HH241A-D). HH240A-D appear as a series of molecular bow shocks, similar to those seen in, for example, the Cepheus A outflow. However, unlike CepA, where the bows are clustered in a chaotic group, in L1634 the bow shocks appear periodically along the outflow axis. Indeed, these data suggest that the L1634 outflow may be variable, and particularly that the H<sub>2</sub> bows may be the result of a ``pulse'' jet.

Pulsed jets generate internal working surfaces along the jet beam, where faster sections of the jet catch up with and shock slower sections. Each working surface consists of a layer of dense gas that is bounded both upwind and downwind by a radiative shock. Spillage of this shocked gas in a direction perpendicular to the flow axis, and the subsequent ``sweeping back'' of the gas as it interacts with the ambient medium, will result in a bow shaped configuration for each leading, working surface shock. These may appear much like the H₂ bow shocks in L1634 (HH240A-D). However, the bow shock sizes are expected to increase with distance from the source (due to the continuous sideways expulsion of material); for HH240A to HH240D, the H₂ data suggest that the bows decrease in size with distance from the source. Nevertheless, the spacing between the HH240A-D bow shocks, and the distance to the first bow shock working surface, do infer a realistic period for the velocity variation, of 300--500 years (assuming a jet velocity of 100 km s^-1 and a velocity variation of 10 km s^-1). Moreover, recent modeling of molecular jets do result in emission maps that are qualitatively similar to L1634, particularly if the jet is variable.

It seems likely, then, that the well-defined bow shocks HH240A-D in the L1634 outflow are best explained with a variable jet model. Indeed, L1634 is perhaps the best example to date of a pulsed, molecular jet. Spectroscopic data and proper motion studies would, however, allow us to better understand these remarkable molecular bow shocks.

Detection of H3+ in Molecular Clouds Confirms Theories of Interstellar Gas Phase Chemistry

Tom Geballe

Head of UKIRT Operations

Takeshi Oka, B.J McCall

University of Chicago

UKIRT and CGS4 recently have detected the presence in molecular clouds of the H₃⁺ molecular ion (see Figure 1), whose existence has been regarded for nearly a quarter of a century as a cornerstone of interstellar chemistry. H₃⁺ is a stable arrangement of three protons and two electrons, which is relatively easy to make in the laboratory (via the reaction of H₂⁺ with H₂). Since the detection in 1980 of its infrared spectrum in the laboratory, H₃⁺ had been sought unsuccessfully in interstellar space. Meanwhile, its emission spectrum has been discovered in the ionospheres of Jupiter, Uranus, and Saturn. UKIRT has figured prominently in these discoveries and in several subsequent detailed studies (eg, Oka and Geballe, 1990, ApJ 351, L53; Trafton et al, 1993, ApJ, 405, 761; Geballe et al, 1993, ApJ, 408, L109; Lam et al., 1997, Icarus, in press).

Production of H₃⁺ in the giant outer planets is precipitated both by collisions of charged particles from the solar wind with the H₂-rich magnetospheres of the outer planets and by dissociation of H2 by solar UV. In interstellar clouds H₃⁺ is predicted to be produced following collisions of cosmic ray particles with hydrogen molecules. A highly reactive species, H₃⁺ is believed to initiate chains of chemical reactions, which result in its destruction, but ultimately in the production of many of the polyatomic molecules observed by astronomers in interstellar clouds. Indeed it has not been possible to explain the existences and abundances of many of these molecules without invoking the presence of H₃⁺ However, due to its very low abundance in interstellar clouds, until last year H₃⁺ had remained undetected in space.

Spectra of GL2136 obtained in April and July 1996. A pair of H₃⁺ lines, from the lowest lying ortho and para levels, are indicated by vertical arrows. The wavelengths of the lines in the two spectra are shifted due to the earth's orbital motion. The emission feature in the April spectrum is due to incomplete cancellation of a strong complex of telluric methane lines.

This group were among several which had previously failed to detect H₃⁺ in interstellar space (e.g. Geballe & Oka, 1989, ApJ, 342, 855). The improved sensitivity of CGS4 following installation of its 256x256 detector array interested us in reviving our search. In the spring and summer of 1996 we succeeded in identifying the faint spectral signature of H₃⁺ toward two highly embedded young stellar objects, AFGL2136 and W33A (1996, Nature, 384, 334). More time was awarded by PATT in Semester 97B (February and July) to extend the search resulting in detections of H₃⁺ in at least two additional embedded IR sources, out of seven observed

The amounts of H₃⁺ detected in the interstellar clouds approximately match predictions based on the expected rates of production of H₃⁺ via cosmic rays and destruction via chemical reactions. Thus, the results support theories of interstellar chemistry first proposed in 1973 (Herbst and Klemperer, 1973, ApJ, 185, 505; Watson, 1973, ApJ, 183, L17). Detailed comparisons of the abundance of H₃⁺ and other key molecules will allow them to accurately determine the role that reactions involving H₃⁺ and its descendants play in the chemical evolution of molecular clouds.

---

Previous: UKIRT Highlights | Up: UKIRT Newsletter | Next : UKIRT News