The first cooling flow cluster with a central radio source observed
by Chandra was Hydra A
(McNamara et al. 2000;
David et al. 2001,
Nulsen et al. 2002).
Hydra A is a powerful, doubled-lobed, FR I radio source
(3C 218), and the radio lobes were
found to be anti-correlated with the cluster gas. The cavities evacuated
by the radio source are approximately 25 kpc in diameter. An image
of the radio source / X-ray gas interaction is shown in
Figure 1, and is from
McNamara et al. (2000).
The cooling time in the center of the cluster is 6 × 108
yr, and surprisingly, the coolest gas was found in the regions
surrounding the lobes. This is seen in the temperature map from
Nulsen et al. (2002),
shown here in Figure 2. There is currently no
evidence that the radio source is strongly shocking the ICM, but weak
shocks are not ruled out with a limit on the Mach number of
< 1.23.
The cooling of the ICM is found to occur only over a limited temperature
range, and repeated outbursts from the radio source would be necessary to
prevent cooling to lower temperatures
(David et al. 2001).
 |
Figure 1. The first radio source / cluster cooling flow interaction observed with Chandra: Hydra A (McNamara et al. 2000). |
 |
Figure 2. The temperature map of the Hydra A cluster, with radio contours superposed (Nulsen et al. 2002). The coolest gas is found in the regions surrounding the radio source, and there is no evidence for current strong-shock heating of the ICM from the radio lobes. |
The Perseus cluster (Abell 426) was first observed with Chandra in early 2000 (Fabian et al. 2000). This nearby cooling flow cluster (z = 0.0183) is the brightest cluster in the X-ray sky and contains the powerful double-lobed radio source 3C 84 in the central galaxy, NGC 1275. Perseus provides one of the clearest examples of the interaction between the central radio source and the X-ray-emitting ICM, and signs of the interaction were already seen in the ROSAT data (Böhringer et al. 1993). The cluster exhibits two distinct bubbles in the X-ray gas that are filled with 1.4 GHz radio plasma and surrounded by bright shells of X-ray emission. The adaptively-smoothed Chandra image of the center of Perseus, with radio contours superposed, is shown in Fig. 3 (Fabian et al. 2000). The cooling time in the cluster center is approximately 108 yr, and no evidence for current strong-shock heating is seen in the temperature map (Fig. 4, Schmidt et al. 2002). The regions surrounding the radio source are the coolest in the X-ray. Presented at this meeting were initial exciting results from a much deeper observation of the Perseus cluster that was performed by Chandra in 2003, revealing intriguing ripple features in the X-ray surface brightness that are interpreted as resulting from the propagation of weak shocks and viscously-dissipating sound waves into the ICM and resulting from repeated outbursts of the central radio source (Fabian, this conference; Fabian et al. 2003).
 |
Figure 3. Overlay of radio contours onto the adaptively-smoothed Chandra X-ray image of the Perseus cluster. Radio:NSF/AURA/VLA; X-ray:NASA/IoA/A.Fabian et al. |
 |
Figure 4. Temperature map of the central regions of the Perseus cluster (Schmidt et al. 2002). |
Abell 2052 shows structures in its center that are
very similar to those in Perseus. This cluster is nearby (z = 0.0348),
and the central cD galaxy
is host to the powerful, double-lobed FR I radio source 3C 317. The Chandra image
(Blanton et al. 2001,
2003)
revealed clear deficits in
the X-ray emission to the north and south of the cluster center that are
filled with 1.4 GHz radio emission (Fig. 5). The
bubbles are approximately 20 kpc in diameter and are
surrounded by bright shells of X-ray emission. Extrapolating
the smooth density profile of the cluster outside of the shells into the
center of the cluster shows that the mass of gas that would have filled the
holes is consistent with the mass of gas currently found in the shells. This
confirms the visual impression that the ICM has been swept aside by the
radio lobes and compressed into the X-ray-bright shells. The
temperature map (Fig. 6) reveals that the shells
surrounding the radio lobes are cool, and show
no evidence of being strongly shocked, with a limit to the Mach number of
< 1.2. The cooling time
in the shells is 2.6 × 108 yr.
An overlay of optical emission line contours of
H
+ [N II]
onto the X-ray image (Fig. 7) reveals a striking
positive correlation between the brightest parts of the X-ray shells and
the optical line emission. The optical emission represents gas with a
temperature of
104 K, and
the temperature of the X-ray gas in these same regions is
107 K, so
at least some gas is cooling to low temperatures in this cooling flow
cluster. The cooling time of the shells is approximately an order of
magnitude longer than the radio source lifetime of 107 yr,
and therefore, the majority of the cooling to
low temperatures in the X-ray shells likely occurred when the gas was closer
to the cluster center and was subsequently pushed outward by the radio
source.
 |
Figure 5. Overlay of 1.4 GHz radio contours onto the adaptively-smoothed Chandra image of Abell 2052 (Blanton et al. 2001, 2003). |
 |
Figure 6. Temperature map of the central region of Abell 2052 derived from the Chandra data (Blanton et al. 2003). The shells surrounding the radio source are cool and show no signs of strong-shock heating. |
 |
Figure 7. Overlay of optical emission line contours from Baum et al. (1988) onto the adaptively-smoothed Chandra image of the center of Abell 2052 (Blanton et al. 2001). |
Another example of radio source / X-ray gas interaction in a cooling flow cluster observed with Chandra is Abell 262 (Blanton et al., in preparation). This cluster is at a redshift of z = 0.0163 in the same supercluster as Perseus, but is less luminous than Perseus in the X-ray, and has a smaller cooling flow. The double-lobed radio source is orders of magnitude less powerful than those in the previous examples, with a power at 1.4 GHz of only P1.4 = 4.7 × 1022 W Hz-1 (Parma et al. 1986). Still, the radio source has blown a bubble in the ICM, clearly seen to the east of the cluster center (Fig. 8). This bubble is much smaller than those in the other clusters discussed above, with a diameter of 5 kpc compared to 20 - 25 kpc for the others. Similar to the other cases, the cooling time for the gas surrounding the radio lobes is 3 × 108 yr, and the temperature map shows no evidence of strong-shock heating (Fig. 9).
 |
Figure 8. Overlay of 1.4 GHz radio contours (Parma et al. 1986) onto the adaptively-smoothed Chandra image of Abell 262 (Blanton et al., in preparation). |
 |
Figure 9. Temperature map of the central region of Abell 262 (Blanton et al., in preparation). |