2002ApJ...564..688R

We used a direct CCD camera at the Magellan I telescope at Las Campanas Observatory and the Focal Reducer/Low Dispersion Spectrograph (FORS1) at the Antu Very Large Telescope (VLT) ESO Paranal Observatory to image fields centered on the inner and outer optical filaments in the halo of NGC 5128. In the V versus U-V color-magnitude diagrams we have identified young blue supergiants associated with these line-emitting filaments located between the inner radio lobe and the northern middle lobe. Around the outer filament, stars as young as 10 Myr were detected. They are principally aligned with the direction of the radio jet, but a vertical north-south alignment along the edge of the H I cloud is also present. Young stars in the inner filament field are found inside the bright knots of photoionized gas and are strongly aligned in the direction of the center of the galaxy at the same position angle as the inner radio jet. Fitting the Padova isochrones on UV color-magnitude diagrams, we find that blue stars around the inner filaments have ages similar to the ones around the outer filaments ∼10-15 Myr and the same abundance of Z=0.004. The presence of young blue supergiants clearly shows that the bright blue knots in the northeastern halo of NGC 5128 are associations of young stars with photoionized gas. The temperature of the brightest stars is T∼12,000-16,000 K, insufficient to account alone for the high excitation lines observed in the surrounding ionized gas. Thus, the optical emission jet is principally seen due to its alignment with the radio structure of the active galactic nucleus (AGN). The highly collimated star formation is present only in the northeastern halo of the galaxy, suggesting interaction of the jet with the gas clouds deposited during the last accretion event as the preferred triggering mechanism. From these observations, we infer a lower limit for the age of the NGC 5128 jet at 10⁷ yr. The triggering of the star formation in the dense clouds in the halo of the galaxy by the jet supports the alignment effect observed in high-redshift radio galaxies. It also suggests that radio galaxies should have higher than normal star formation rates.