A new detection of shock heating in NGC 3801

Our recent Chandra observations of NGC 3801 [13] revealed a second definite example of strong shocks produced by a small FRI source on galaxy scales. Fig shows the Chandra-detected emission from NGC 3801, which traces well the outer edges of the radio lobes.

We can rule out a non-thermal model for the X-ray emission based on its spectrum, and find best-fitting mekal temperatures of 1.0 keV and 0.7


Fig. 1. The shell of shocked gas around the south-western lobe of Centaurus A as observed with Chandra [2, 10]. 20cm radio contours [17] are overlaid.


Fig. 1. The shell of shocked gas around the south-western lobe of Centaurus A as observed with Chandra [2, 10]. 20cm radio contours [17] are overlaid.

keV for the West and East lobes, respectively. The undisturbed interstellar medium has a temperature of 0.23 keV. We find that the observed density contrast is consistent with the value of 4 expected for a strong shock, using the mean properties of the shell and the ISM density halfway along the lobe. The shells are overpressured by a factor of 13 - 20 and the shell pressure is ~ 7 times the synchrotron minimum internal lobe pressure (consistent with the general finding that FRI minimum pressures are typically an order of magnitude lower than external pressure [14, 15, 16]).

We estimated the shock Mach number using two methods, as descibed in more detail in [13]: applying the Rankine-Hugoniot jump conditions using the observed temperature jump gives M^ 3 — 4; alternatively, ram pressure balance gives M ~ 5 — 6. The discrepancy between the two methods is probably due to the expected temperature and density structure of the shell and the interstellar medium (in both cases the data are too poor to constrain

Fig. 2. The shocked gas shells of NGC3801 as observed by Chandra. Top: binned 0.5 - 2 keV counts; bottom: Gaussian smoothed 0.5 - 2 keV image with 20 cm radio contours overlaid from new VLA data.

these). Nevertheless this is a clear detection of strongly shocked gas with M ^3 — 6, which implies a lobe expansion speed of ~ 600 — 1200 km s_1.

The total thermal energy stored in the hot gas shells is ~ 8 x 1055 ergs, and for M^ 4, the total kinetic energy of the shells is ~ 9 x 1055 ergs. The total energy of the shells, 1.7 x 1056 ergs, is comparable to PintV, the approximate total energy available from the radio source as work; however, it is ~ 25 times the minimum work required to inflate the lobe cavities 7 x 1054 ergs), so that a simple calculation of the radio-source energy input from the cavity size would be a significant underestimate. The total energy is also equivalent to the thermal energy of the ISM within 11 kpc (or 25 percent of the thermal energy within 30 kpc). Shock heating is therefore the dominant energy transfer mechanism during this phase of radio-source activity, and will have dramatic long term effects: part or all of the ISM may be expelled from the galaxy, and the entropy of the gas will be permanently increased. The internal energy of the radio source 4 x 1056 ergs) must also eventually be transferred to the environment.

The age of the radio source in NGC 3801 is estimated to be ~ 2 x 106 y from radio spectral ageing and dynamical arguments, which implies an energy injection rate of ~ 3 x 1042 ergs s_1. This should correspond to a considerable fraction of the jet power, which is consistent with a rough estimate of its jet power based on scaling that of 3C 31 [18] by the ratio of radio luminosities of NGC 3801 and 3C 31. The rate of mechanical energy extracted is roughly an order of magnitude higher than the accretion-related X-ray luminosity, so that the AGN is more efficiently converting energy into jet production than radiation. We also find that the Bondi accretion rate from hot gas would be sufficient to power this radio outburst, for n ~ 0.05.

0 0

Post a comment