2. Properties of the group of galaxies and extended narrow- emission line nebulosity
3C 275.1 (1241+166) was the first quasar found at the centre of rich cluster of galaxies (Hintzen and Stocke 1986). The galaxy field around 3C 275.1 has been investigated by Hintzen et al. (1981), Hintzen (1984) and Ellingson and Yee (1994). They have found an excess of galaxies around it and have measured the redshifts of the number of galaxies. We have used these redshifts to discuss the dynamics of galaxies in the vicinity of 3C 275.1. The comparison of redshifts estimated by Hintzen and Ellingson and Yee indicates that they agree well with the exception of 3C 275.1-2=381. Since Hintzen (1984) determined redshift of this galaxy using four spectral lines, while Ellingson and Yee only from one line, we have taken his redshift. For all galaxies we have calculated the relativistic radial velocities . To remove non-cluster galaxies we have applied the "3- " procedure described by Yahil and Vidal (1977). Accordingly, all galaxies with velocities more than away from the central velocity have been rejected. The process was repeated until we obtain a stable number of galaxies, namely 6 galaxies given in Table 1.
Table 1. Dynamical data for group of galaxies around 3C 275.1
These galaxies belong to a group dynamically bound to the quasar. We have calculated the relativistic radial velocity of the group and the velocity dispersion , which are and respectively. Assuming a radius for the group of galaxies equal to , which corresponds to , and the optical position of the host galaxy of 3C 275.1 to be the group center, we derived the distances of the galaxies from this center "d", the distances in respect to the group radius " " and the position angles measured from the north towards the east "PA" .
In Table 1 the names of galaxies, their redshifts, the calculated relativistic radial velocities, the relativistically corrected differences of radial velocities, the distances "R" and the projected distances " " from the core (see Sect. 3) are given. The relative errors of radial velocities estimated from the measured accuracy of redshifts are smaller than 0.14 %. We find that the host galaxy of 3C 275.1 with its (see Table 1) lies at the centre of the gravitational well of the group.
We have also applied the biweight estimators described by Beers et al. (1990). Thereafter, we have determined the biweight location and the scale for galaxy velocities in the group. It is known that and are good estimators for Gaussian as well as non-Gaussian distributions. They become superior when only few galaxies are involved as in the case of 3C 275.1. We have derived 124378 and 529 for and respectively. Therefore, agrees with within the standard deviation. That confirms our earlier suggestion that the host galaxy of 3C 275.1 lies at the bottom of gravitational potential well of the group.
Recently, Rector et al. (1996) discussed the relation between the quasar-galaxy spatial covariance amplitudes and the radio morphological characteristics of radio sources in clusters of galaxies including 3C 275.1. In the sample analyzed, for the field of galaxies around 3C 275.1 has the highest value, i.e. , which contradicts the six galaxies considered by us as members of the group. The whole sample discussed by Ellingson et al. (1994) consists of 71 galaxies, which might be members of the group of galaxies around 3C 275.1 on the basis of the commonly used criterion. Here is the apparent brightness of the third bright galaxy in the cluster. Our analysis confirms how misleading such a membership criterion may be, particularly for distant clusters of galaxies. We have discovered a foreground group of galaxies at containing 40 galaxies, which is not dynamically bound with the group around 3C 275.1. On the other hand, the amplitude of the angular covariance function is only 0.60 for the whole field, namely for the sphere of radius . The amplitudes of the spatial correlation function are always 100.
We derived the cluster dynamical time, defined as , and the galactic dynamical time () of the order of yrs and yrs respectively for kpc and . Moreover, there is 381 galaxy at a distance of only from the quasar. It lies along the direction (within the possible estimation error) of the NW lobe and the core of 3C 275.1. The impact of the radio jet on its halo might result in the formation of an oblique shock and cause the non-collinearity of the radio structure. We shall discuss this in Sect. 3.
The host galaxy of 3C 275.1 () is similar to cD galaxies in nearby clusters of galaxies. The elliptical nebulosity, which is centered on the quasar, extends from the north- east towards the south-west at position angle about and measures (i.e. 104 52 kpc) in the R-photometric system. In the V - photometric system only the stellar nucleus of the QSO is seen.
Now, we shall use the X-ray observations of 3C 275.1 and narrow-emission line measurements to determine the particle number densities. The Einstein IPC observation indicated that the QSO, group, or both are strong X-ray sources: in the keV band (Tananbaum et al., 1979). Since the X-ray source is marginally resolved, in size, the observed X-ray flux may be due to either the quasar itself, to a cooling flow, or both. Based upon the PSPC observations with the ROSAT in the observed keV band ( keV in the rest frame of QSO), Arp (1996) fitted a Raymond-Smith spectrum of temperature K and Galactic neutral hydrogen column density of as well as a simple power-law with an energy photon index modified at low energy by photo-electric absorption in a uniform column density of cold gas.
On the other hand, from the HI observations at 21 cm by Stark et al. (1992) we estimate , which is intermediate between two above quoted values. The accuracy of the observations leads to between 1.3 and . Therefore, it seems that above mentioned three values of hydrogen number density differ only marginally. It is worth noting that measured X-ray emission is significantly extended (about ) and might be emitted by hot gas of the cooling flow. Furthermore, an upper limit for the temperature of the ICM gas can be derived from the equilibrium condition between the kinetic energy of the group of galaxies and the thermal gas energy, i.e. (here g is the proton mass, -the mean molecular weight). Simulations of X-ray clusters by Navarro et al. (1995) indicate that this condition is fulfilled moderately well by the " -model" over all but the innermost regions. Putting for the solar abundances (Edge & Stewart, 1991) and the velocity dispersion of the group of galaxies (see Table 1), we derived K, while for the upper limit for the gas temperature is about K. In further calculations we take for . Such a cool ( K) gas is characteristic of cooling flows. The calculated temperature refers to the whole group of galaxies, while the temperature K might occur in the central region of the group, where cooling flow appears. In the standard picture of cooling flows (Fabian et al., 1984, 1991; Sarazin, 1986) as the hot gas cools through K and condenses, it will emit EUV and soft X-ray radiation, which can photo-ionize the previously condensed cooler denser matter and fuel an optical emission-line spectrum.
We derive the lower limit on the number density of the ICM gas from the condition that the cooling time should be shorter than the Hubble time . The present Hubble time is given by . The Hubble time at the redshift of 3C 275.1 is yrs, while the cooling time scale is defined as (where n is the number density and -the standard cooling function or the volume emissivity). In the temperature range K, the numerical calculations indicate (Raymond & Smith, 1977). Therefore, the lower limit on the number density of ICM gas is given by: For K and we have calculated .
It seems that accreting matter from a cluster cooling flow of the type seen in low-redshift clusters, which are X-ray sources (Stewart et al., 1984) is the material in the nebulosity. Hutchings (1992) found that extended emission structure occurs in all RLQSOs discussed with a variety of properties and z from 0.36 to 0.91. However, it is generally stronger in the large lobe-dominated (steep spectrum) sources, such as 3C 275.1, than in compact (flat spectrum) ones.
Crawford and Fabian (1989) have obtained the narrow-band images of 3C 275.1 along the major axis of the elliptical nebulosity centered on the quasar in the O[III] nm and [OII] nm emission lines redshifted to 778.6 nm and 579.6 nm respectively as well as in a nearby line-free continuum band. They have found that the narrow emission lines of [OII] and [OIII] extend well into the nebulosity on either side of the nucleus up to in NE and about in SW directions. The upper limit on the number density of the cooler extended and denser emission line gas is determined by the pressure equilibrium with the ICM gas (Stockton & Mac Kenty 1987). In the other case, the cooler gas having K, which should be accurate to a factor of 2, contained within the radius of 52 kpc will disperse at its internal sound speed , i.e. about during the acoustic time scale , i.e. yrs, where "r" is the radius of the ENLR in kpc. The acoustic time scale is comparable with the Hubble time for z=0.5549. Therefore, from the pressure equilibrium condition we obtain , which gives () for K and K. The cooling time will be shorter and number density greater for central regions of the nebulosity.
However, Hintzen and Romanishin (1986) found that the narrow-emission nebula displays considerable structure. There are several knots each with luminosity of , while the total [OII] luminosity is about (Hes et al., 1993). Hence, inhomogeneous cooling might occur in 3C 275.1 similar to that observed in some nearby clusters of galaxies (Prestwich et al., 1995). We shall calculate the particle number density in such knots. The condensates will cool isobarically, while the dynamical time is smaller than the cooling one, which is fulfilled. As a dynamical time we consider the time required for a sound wave to cross the cooling cloud, i.e. . Here P is the pressure of the cloud gas and -the mean mass per particle. For the mean molecular weight above mentioned , we have . This condition implies an upper limit on the column density of particles in such a condensate/or cloud, namely , which gives about . However, some clouds will evaporate. The thermal condition and the limitation on the clouds which survive are , where is the standard conduction coefficient given by (Böhringer & Fabian, 1989). Putting K we calculate . Therefore, the condensing clouds might contain particle number densities between and , namely about particles per . On the other hand, the critical density of the particle will be determined by selfgravitation. The condensations will survive, when in the potential well, i.e. roughly when the inflow velocity approaches the sound velocity. Here, is the free-fall time. For K, we have , and thermal instabilities ought to develop. The gas will continue to flow into the central region and condense until self-gravitation induces clouds to coalesce and collapse. Self-gravitation will begin to dominate cloud dynamics at a total density larger than regardless of the temperature of the inter-cloud gas. Here is the pressure of intra-cloud gas given in scale and -mean mass per particle. Putting we derived and the critical number density (), which is about a factor 6 larger than calculated above. Hence, self-gravitation will be insignificant for narrow-emission line clouds of 3C 275.1. Pinkney et al. (1996) and Wilson (1992) discussed the connections between the radio structure and the emission in forbidden-narrow lines. Jets might also be deflected by clouds in ENLR as previously mentioned. The gas emitting in the outskirts of radio lobes could form an irregular "Faraday screen" and be responsible for depolarization of the radio emission. Such depolarization is observed, especially in the SE lobe.
The narrow-emission line spectra are typically reproducible by photo-ionization by a relatively dilute power-law continuum. Hence, knowing the changes of the narrow-emission line ratios, namely [OIII] nm/[OII] nm, with distance from the nucleus and assuming photo-ionization by the QSO core, the pressure profiles of the gas can be deduced (Crawford and Fabian 1989). However, to estimate the ionizing luminosity, the X-ray luminosity of the QSO must be known. To derive the minimum pressure for the line emitting gas a power law between the optical and X-ray wave-bands has been assumed and the spectral index has been taken from Worrall et al.,1987. Hence, the power-law luminosity is , which is a little lower than is typical for the " " QSOs. In Fig. 1 the pressure profiles of the ionized gas for 3C 275.1 and 3C281 (z=0.599) (Crawford et al., 1991)-a QSO almost at the same distance as 3C 275.1 and also lying in a group of galaxies (Yee & Green, 1987) are compared with changes of the pressure of the ICM gas. The changes of the pressure of the ICM gas with distance are described by " -model, i.e. for smaller distances and for distant regions of the cluster. Here and is the -parameter. The central particle densities have been calculated from the X-ray luminosities and core radii as given by . We put and two values for , namely and . The former one represents the typical core radius of cluster of galaxies, while latter refers to the radius of cooling flow region. Hence, we have , and ; for K. The calculated gas pressure profile of ICM gas is shown in Fig. 1.
It is seen that the pressure of gas close to 3C 275.1 is lower than that around 3C281, which has a mass deposition rate (Bremer et al., 1992). In general, the derived pressure profiles of the extended emission-line gas toward the north-east of 3C 275.1 are similar to those of groups of galaxies, e.g. MKW 4 or MKW 3s, and are lower than those of rich clusters of galaxies. However, extra ionization due to interaction with the radio plasma might be invoked especially for the SW side.
The rate of condensation of gas extending over a region of radius at the center of a cluster is releasing a luminosity , where K and . Putting (7.44 kpc), K and we calculate a condensation rate of about . The integrated mass deposition rates or condensation rates should be regarded as uncertain by . Therefore, 3C 275.1 lies in a moderate cooling flow. Although no simple correlation between , and optical line luminosity exists, Allen et al. (1995) showed that the most optically line-luminous clusters contain large cooling flows and exhibit short central cooling times of the order of a few yrs.
© European Southern Observatory (ESO) 1998
Online publication: May 15, 1998