The quasar jets survey

Introduction

Before the launch of Chandra X-Ray Observatory on 23 July 1999 our knowledge of X-ray jets was limited to only a handful of the nearest and brightest active galaxies, namely Centaurus A, M87, 3C120 and 3C273. However the huge improvement in angular resolution and sensitivity provided by Chandra has enabled the discovery of X-ray jets in abundance.

During its in-orbit commissioning phase, Chandra was turned toward the bright, southern quasar PKS 0637-752, which was expected to be a point source and thus a good target for focusing its highly polished mirrors. To everyone's surprise these observations resulted in the serendipitous discovery of a new X-ray jet (Schwartz et al. 2000; Chartas et al. 2000). This article reviews what we have learnt about PKS 0637-752 so far, describes a search for X-ray jets with Chandra and the important contribution being made by the new millimetre-system at the Australia Telescope Compact Array.

The PKS 0637-752 story

Compact Array 3 and 6-cm images of PKS 0637-752 show a bright westward-pointing jet coincident with the X-ray emission discovered by Chandra. The straight jet ends in a northward bend past which no X-rays are detected. It is also at this bend where the radio polarisation electric vectors begin to rotate from perpendicular to the jet to parallel. Archival HST data revealed three previously overlooked unresolved features at 27th magnitude coincident with the brightest parts of the X-ray and radio jets (Figure 1). These observations have allowed the emission mechanism of the jet to be investigated. The observed Spectral Energy Distribution (SED) of one of the jet features is shown in Figure 2. While the radio jet is undeniably due to synchrotron radiation because of the high linear polarisation, the emission mechanism of the X-ray jet is less clear.

Figure 2: The spectral energy distribution of one of the knots in the jet of PKS 0637-752.

The Compact Array polarisation data reveal that the magnetic field in the jet does not show Faraday rotation, implying that the X-ray emission cannot be thermal. If the X-rays are synchrotron then a simple power-law between the radio and X-ray flux densities over-predicts the optical flux density by two orders of magnitude. A more probable interpretation is that the X-rays are the result of inverse Compton scattering of photons by the electrons responsible for the synchrotron radiation. The most likely source of these photons originating so far, 100 kpc, from the quasar core is the cosmic microwave background. However this still requires the kpc-scale jet to be relativistic with a bulk Lorentz factor of ~ 20 and a small angle to the line-of-sight (Celotti, Ghisellini & Chiaberge 2001). Further, the X-ray intensity of the three bright knots falls off more quickly with increasing distance from the core than in the radio which is not predicted by this model. Neither does the model explain why the X-ray emission ceases at the bend.

Clearly more work is required on modelling and the high-resolution observations of the radio jet, now available as a result of the Compact Array millimetre upgrade, can reveal sub-clumps etc, and are crucial for model refinement and furthering physical understanding of the jet.

The quasar jets survey

This discovery of the X-ray jet in PKS 0637-752 poses some interesting questions:

* How common are X-ray jets among high-power AGN?
* What conditions are required to produce them?
* What role does jet bending and polarisation play?

To help answer these questions we are undertaking an X-ray survey with Chandra of a sample of AGN with arcsec-scale radio jets. The target sources are drawn from southern hemisphere Compact Array and northern hemisphere VLA imaging surveys of flat spectrum radio sources (Lovell 1997; Murphy, Browne & Perley 1993). Sources were selected from these samples if they have emission on scales greater than 2 arcsec, allowing any X-ray jet emission to be clearly distinguished from the core. Of these sources, two sub-samples were selected for observation by Chandra:

A. Based on the predicted X-ray jet count rate. This was done assuming the radio to X-ray flux ratio of our sources was the same as PKS 0637-752. Sources were chosen if more than 30 X-ray counts were predicted.

B. Based solely on jet morphology to ensure that X-ray properties for a variety of jet types will be examined.

There are 56 AGN in the sample with a large overlap between the A and B sub-samples.

Figure 3: A selection of sources with new X-ray detections from our survey. Compact Array 8.6-GHz data are shown as contours and X-ray data convolved to the same resolution are shown as pixels. From left to right the sources are PKS 0208-512, PKS 0920-397, PKS 1030-357 and PKS 1202-252.

To date 20 targets have been observed by Chandra resulting in 12 new X-ray jet detections which are now the subject of detailed, multi-wavelength investigation. We find that the X-ray morphology closely follows the radio with sharp bends marking the termination of X-rays only in some cases (Figure 3). It seems that the FR-II X-ray jets can be interpreted in the same way as PKS 0637-752: as inverse Compton scattering of the Cosmic Microwave background (Marshall et al, 2004).

PKS 0637-752 revisited

The millimetre upgrade to the Compact Array is providing the opportunity to image these objects at higher resolution which will allow a more detailed investigation of the jets and their emission mechanism. In May 2004 we conducted 12-mm observations of the objects in our sample already observed with Chandra including our old friend PKS 0637-752. The 20.1-GHz image is shown in Figure 4 and reveals the kpc-scale radio jet in unprecedented detail. The gentle curve of the jet as it leaves the core is clearly seen and individual jet knots are resolved. Several of these knots in the inner jet do not appear to have clear X-ray counterparts despite the presence of overlying X-ray emission. This may be an effect of different angular resolutions of the Compact Array and Chandra images with the X-ray image showing blended emission from multiple knots, or it may be indicating significant differences in the locations of knots. We will use the new Compact Array data to model the SEDs of the knots which we hope will help understand the changing conditions along the jet and the cause of the cessation of X-rays past the northward bend.

Figure 4: A string of pearls. Our recently obtained 20.1-GHz Compact Array image of PKS 0637-752 (pixels) overlayed with contours of X-ray emission.

In some cases the jets will be bright enough to be detected by the Compact Array at 3 mm and the presence of a bright compact core provides a handy in-beam phase reference.

Lastly it is interesting to note the roles played by the Compact Array and Chandra data in understanding the electron energy distribution of the jet. Under the inverse Compton interpretation the X-ray data provide information on the low energy electrons while the radio data probe the high energy end of the spectrum. Thus it is the radio emission rather than the X-ray that is revealing new information on the highest energy phenomena in quasar jets!

References

Celotti, A., Ghisellini, G. & Chiaberge, M. 2001, MNRAS, 321, L1
Chartas, G. et al. 2000 ApJ 542 655
Lovell. J. 1997, PhD Thesis, University of Tasmania
Marshall, H. L. et al. 2004 ApJ Supp, accepted
Murphy, D. W., Browne, I. W. A. & Perley, R. A. 1993 MNRAS, 264, 298
Schwartz, D. A. et al. 2000, ApJ, 540, L69

Jim Lovell (ATNF), Leith Godfrey (RSAA), David Jauncey (ATNF), Geoff Bicknell (RSAA),
Mark Birkinshaw (CfA/U. Bristol), Jonathan Gelbord (MIT), Markos Georganopoulos
(U. Maryland), Herman Marshall (MIT), David Murphy (JPL), Roopesh Ojha (ATNF), Eric Perlman (U. Maryland), Dan Schwartz (CfA) and Diana Worrall (CfA/U. Bristol)

(Jim.Lovell@csiro.au)