New constraints on protostellar jet collimation from high-density gas UV tracers

Abstract

The analysis of high-resolution profiles of the semiforbidden UV lines of C III](1908) and Si III](1892) in the spectra of T Tauri stars (TTSs) shows the following : (1) There is C III](1908) and Si III](1892) emission at velocities that are similar to those observed in the optical forbidden lines formed in the TTSs jets. The luminosity of the UV lines is comparable to that of the optical lines. (2) The comparison between the optical and UV light curves indicates that the C III](1908) and Si III](1892) emission of RY Tau is not associated with accretion shocks, but it is produced farther than 2 R-* from the star. (3) The profiles of the UV semiforbidden lines are significantly broader than those of the optical forbidden lines. These profiles cannot be produced in a narrow collimated beam, and they are most likely produced in a bow-shaped shock wave formed at the base of the optical jet, where the hot gas emits in a broad range of projected radial velocities. (4) The atmosphere of RU Lup contributes significantly to the Si III](1892) emission. (5) A puzzling narrow feature is observed close to the C III](1908) line. The feature is blueshifted by -260 km s(-1), which corresponds to the wind terminal velocity measured in the P Cygni profile of the Mg II (UV1) lines. Moreover, constraints are derived on the characteristics of the C III](1908) and Si III](1892) emitting region in RY Tau. It is shown that 4.7 less than or equal to log T-e less than or equal to 5.0 and 10(9) cm(-3) less than or equal to N-e less than or equal to 10(11) cm(-3) provided that the emission is produced in a collisional plasma and that the 1665 Angstrom feature observed in low-dispersion International Ultraviolet Explorer (IUE) spectra is confirmed to be O III](1665) emission produced in the wind. These very high densities are difficult to generate in the shocks produced by the magnetic pinching of centrifugally driven magnetized disk winds. The data also suggest that the shocked layer has a radius of some few stellar radii and it is closer than similar to 38 R-* to the star.