Elliptical galaxies, consisting of up to a trillion (10¹²) stars, have all the characteristics of old objects in the Universe, which have not changed much during the last few billion years. In such steady systems, the motions of stars obey well-established rules, which are a direct consequence of classical mechanics. In particular, if we know the velocity dispersion of stars and the size of the galaxy we can immediately estimate the mass of the galaxy. However, this is not the only way to measure the galaxy mass. Massive elliptical galaxies often posses hot gaseous atmospheres, which are powerful sources of X-ray emission. If the gas is in hydrostatic equilibrium - the galaxy's gravity is balanced by the pressure of the gas, then we can use the gas to make an independent estimate of the galaxy mass. For this we need to measure accurately the gas temperature and its spatial distribution.

With the launch of the Chandra X-ray observatory, having angular resolution of order of 0.5 arc second, the quality of X-ray images is comparable to ground based optical data (Figure 1), and therefore the mass of the galaxy can be equally well determined by each method. Any disagreement between the two methods is a valuable tool to determine the characteristics of the gas. For instance, it has been suggested that relativistic protons - an elusive constituent of cosmic rays - are often mixed with the thermal plasma that we see in X-rays. Due to their large mass, the protons, unlike electrons, do not produce much radiation and we cannot easily detect their presence in the gas. However, if relativistic protons make a substantial contribution to the gas pressure, then the gas spatial distribution will be broadened. The net result would be an error in the mass determination - X-ray analysis is expected to give a lower mass, compared to the mass determined from the optical data.

This comparison was made for a few well-studied elliptical galaxies. To make the comparison less sensitive to observational noise, the gravitational potential, rather than the mass itself, has been calculated and compared as shown in Figure 2. The potentials derived from X-ray and optical data came up remarkably close to each other with the discrepancy amounting to less than 10%. This immediately translates into a similarly small upper limit on the contribution of cosmic rays to the gas pressure. Moreover, this 10% limit is perhaps applicable to other hard to detect effects including departures from hydrostatic equilibrium, gas motions, substantial magnetic fields or incorrect modeling of stellar kinematics. Of course it is possible that different effects have larger amplitudes, but opposite signs so that they cancel each other. However, it would be a remarkable coincidence that the residual discrepancy is so small. The conclusion reached so far is that in the most round and well-behaved systems the gas hydrostatic equilibrium is a good approximation and the contribution of cosmic rays to the gas pressure is small. The next step will be to look at systems that are more complicated and use the X-ray observations to study the characteristics of their stellar populations, which are difficult to measure by other means.

Eugene Churazov, William Forman, Alexey Vikhlinin, Scott Tremaine, Ortwin Gerhard, Christine Jones.

Publications

Eugene Churazov, William Forman, Alexey Vikhlinin, Scott Tremaine, Ortwin Gerhard, Christine Jones, "Non-thermal pressure in M87 and NGC 1399 gas: X-ray vs. optical potential profiles",
2008, submitted to MNRAS, arXiv:0711.4686