Using HST's high resolution on one hand and its infrared capabilities on the other, astronomers have been able to probe the dusty environments of young stellar objects (YSOs) and of stars in their late evolutionary stages with unprecedented detail. While these observations have not revolutionized the field, they have, on one hand, led to a much deeper understanding of the processes involved, and on the other, opened an entirely new set of questions, while producing some of the most spectacular images.
4.1. Outflows and jets from young stellar objects
Molecular clouds are the reservoirs of mass and angular momentum from which stars are born. In the initial phase, dynamical infall occurs. The finite angular momentum of infalling cloud material leads to the formation of an accretion disk. Angular momentum is transported outward in this disk resulting in the accretion of mass (and some angular momentum) onto the central object (e.g., Najita 2000, Shu et al. 1994, Camenzind 1990).
For a long time it has been known that outflows and collimated jets are signposts of stellar birth (see, for example, Königl 1989, Reipurth and Heathcote 1993). In particular, many of the radiative shocks known as Herbig-Haro (HH) objects were found to be associated with highly collimated jets emanating from the vicinity of YSOs (e.g., Reipurth 1999). In some cases, these jets were found to be several parsecs in length (Reipurth et al. 1998). Figure 9 shows a few remarkable HST images of jets from YSOs.
 |
Figure 9. Montage of four images showing
jets from young stars HST/WFPC2. |
In spite of the ubiquity of astrophysical jets, we still lack a comprehensive theory of their acceleration and collimation (see e.g., review by Livio 2000). The most promising mechanism relies on an accretion disk (around a central compact object like a YSO or a black hole) threaded by a large-scale poloidal magnetic field. Some magnetic flux is assumed to be in open field lines, inclined by an angle of more than 30° to the vertical to the disk's surface. Ionized gas is forced to flow along field lines. Since the foot points of these lines are anchored in the disk and rotate with it, material is accelerated centrifugally like beads on rotating wires (Blandford and Payne 1982, Ogilvie and Livio 2001). In this picture, the acceleration basically stops at the Alfvén surface, where the kinetic energy density in the jet becomes comparable to the magnetic energy density. Collimation, however, occurs primarily outside the Alfvén surface. There, the field gets wound up by the rotation, generating toroidal loops. The curvature force exerted by the toroidal field acts in the direction of the rotation axis, thus achieving collimation by "hoop stresses." Alternatively, in a vertical field of the form Bz ~ (r2 / Rin + 1)-1/2 (where Rin is the radius at the accretion disk's inner edge), in which the flux is largest at the outer disk, the field lines have a naturally collimating shape (e.g., Spruit 1996). Physically, poloidal collimation is achieved in this case as the material encounters the high flux in the outer disk (especially if Rdisk ~ RAlfvén).
While this theoretical picture of jets being accelerated and collimated by accretion disks has been largely developed prior to any HST observations, HST has provided the first direct evidence for the fact that jets indeed originate at the centers of disks (at least in the case of YSOs). Observations of the objects HH 30 (Burrows et al. 1996, Figure 10) DG Tau (Fig. 9), and a few other YSOs clearly show the jet emanating from the disk center, and they reveal even the illuminated upper and lower disk surfaces.
 |
Figure 10. HH 30 disk/jet, HST/WFPC2.
Credit: NASA, C. Burrows (STScI), J. Krist (STScI),
K. Stapelfeldt (JPL), and the WFPC2 Science Team. |
Furthermore, the high resolution of HST has enabled the determination of proper motions in YSO jets, from images obtained over a time interval of a few years (e.g., Bally 2003). The highest velocities have been observed along the jet axis. Typically, the velocities are of order 200 to 400 km s-1. Since these are precisely of the order of the Keplerian velocity in the inner disk around a YSO, the proper motion observations provide additional support for the jet formation scenario described above. Incidentally, similar observations of the optical jet in the active galaxy M87 also showed proper motion. In this case at the apparent "superluminal" speed was of 4c-6c (Biretta, Sparks, and Macchetto 1999), again corresponding to the fact that one expects relativistic speeds from the vicinity of a central black hole.
Another related phenomenon that only HST could discover is the existence
of externally irradiated small jets ("microjets";
Bally, O'Dell and
McCaughrean 2000).
Many of these microjets (Fig. 11) are only
about 0.1" wide, and are therefore usually undetectable against the
nebular background in ground-based observations. For irradiated jets,
observations of the
H
surface brightness,
the emission measure
(EM = 
ne2 dx; where ne
is the electron density and x is the linear size of the emitting
region), and the jet width, allow for a determination of the electron
density. Together with the jet speed one can therefore obtain the rate
of mass loss in the jet,
j.
Typically, the irradiated small jets in the Orion
nebula are characterized by
j ~
10-9
M
yr-1, at least an order of magnitude lower than the rates
observed in the long jets associated with Herbig-Haro
objects. Generally, the rate of mass loss in jets from YSOs is found to
be about 1-10% of the rate at which mass is accreted through the disk
onto the young stellar object.
 |
Figure 11. A bright one-sided microjet emerging from a proplyd located below Orion's Bright Bar. Adapted from Bally et al. (2000) |
In addition to the important disk-jet connection, HST observations of YSOs have provided another interesting new element, this time related to planet formation.