5.1. Search for the soft excess in and around clusters
Suzaku has observed about 30 clusters of galaxies by April 2007. Several of these observations were done in order to search for soft excess emission. We describe here the results for Abell 2218 and Abell 2052 in some detail since these two clusters have been studied rather extensively, and then briefly review results from other sources.
5.2. Abell 2218 (z = 0.1756)
Abell 2218 is a bright cluster at z = 0.1756 with an ICM temperature of about 7 keV. The Suzaku observations were carried out on two occasions with a total exposure of 80 ks and the results were published by Takei et al. (2007a). Several conditions are favourable to the detection of a soft excess in this cluster with Suzaku. First, the cluster is known to be undergoing a merger, likely along the line of sight, as indicated by two galaxy concentrations at different redshifts. This picture is supported by X-ray observations from Chandra and by the presence of remarkable arcs due to gravitational lensing. Such a merger suggests that a large-scale filament may be located along the line of sight and that the column density of oxygen contained in the warm filament gas could be high. The second reason is that the XIS energy resolution allows to separate a ~ 100 eV cosmologically redshifted oxygen line from a zero redshift oxygen line. Background data in two regions at about 5° offset from the cluster were also taken to obtain information on the oxygen line intensity of the Galactic component.
The time variation of the soft X-ray flux during the observations was very small, so the effect from the solar wind charge exchange emission was considered to be small. A soft excess was searched for in the outside region of the cluster, i.e. for the region with radius greater than 5' (880 kpc at the source) from the Abell 2218 centre. The observed spectra (see Fig. 4) show structures around 0.5 and 0.6 keV, which are the energies corresponding to redshifted O VII and O VIII lines. However, close examination showed that the hump at 0.5 keV can be caused by the oxygen edge in the XIS filters, and the peak around 0.6 keV can be due to Galactic O VII line emission. This unfortunate situation hampered the original XIS capability of detecting the soft excess associated with the cluster. Takei et al. (2007a) set upper limits for the O VII and O VIII line intensities as shown in their Fig. 1, including all the uncertainties in the instrumental response (e.g. by the enhanced contamination in the XIS filters) and in the Galactic line intensities.
Figure 4. Left: Observed XIS spectrum for the outer region of Abell 2218, with upper limits for the redshifted O VII and O VIII lines, for the BI (black) and FI (red) sensors. The background spectra are shown with dashed lines. This figure is taken from Takei et al. (2007a). Right: Observed XIS spectrum for r = 12' - 15'. Spectral fit including Galactic emission (0.1 and 0.3 keV in light and dark blue), cosmic X-ray background (green), and thermal ICM emission with kT = 2.8 keV (red) requires an additional thermal Bremsstrahlung component with kT = 0.7 keV (magenta) whose intensity is constant over r = 0 - 20'. The summed up model (orange) is fitted to the data. Observations with Suzaku confirm the presence of soft excess emission in Sérsic 159-03 and its derived flux is consistent with the values determined with XMM-Newton (Sect. 4.1.3, Werner et al. 2007). However, Suzaku does not confirm the presence of the redshifted O VII lines in the cluster. The excess emission can be fit statistically equally well with a thermal model with low oxygen abundance (< 0.15 Solar) and with a non-thermal model. The figure is taken from Tamura et al. (2007).
Even though the observing conditions were not optimal, the upper limits obtained are nearly an order of magnitude lower than the soft excess level reported in other clusters by Kaastra et al. (2003). With this upper limit, the density of the warm gas can be constrained. The limit for the gas density is:
where H = X b crit (1 + z)3 / mp = 1.77 × 10-7 (1 + z)3 cm-3 is the mean hydrogen density in the universe, where X = 0.71 is the hydrogen to total baryon mass ratio, b = 0.0457 is the baryon density of the universe, crit = 9.21 × 10-30 g cm-3 is the critical density of the universe, and mp is the proton mass. Even though this level of gas density is much higher than the expected density ( ~ 10) in the filaments, the result demonstrates that Suzaku can give a reasonable constraint on the soft excess emission.
5.3. Abell 2052
Suzaku carried out observations of Abell 2052 with 4 offset pointings to cover the cluster outskirts to about 20' (830 kpc at the source). Since the observations were all carried out only 40 days after the launch, the contamination in the XIS filter was not so much a problem. At the carbon K-edge energy, the reduction of the transmission efficiency was less than 8% (see Koyama et al. 2007).
The data were analysed by Tamura et al. (2007). By assuming a single temperature for the ICM and two temperatures (0.2 and 0.6 keV) for the Galactic component, they found a significant soft excess, with a spectrum well described by a featureless continuum modelled (for convenience reasons only) by pure thermal Bremsstrahlung with a temperature around 0.7 keV. The intensity of this soft component is consistent with a constant value over the entire observed field and stronger than the Galactic emission below 0.5 keV. If this emission is associated with Abell 2052, then its luminosity can be comparable to the ICM emission which shows LX ~ 1.4 × 1044 erg s-1 in the 2-10 keV band. The cluster is located fairly close to the extension of the North Polar Spur (NPS), a large spur-like region with strong soft X-ray emission (see e.g. Willingale et al. 2003). However, the Suzaku spectrum of the NPS region showed the temperature to be about 0.3 keV with a fairly strong Mg-K emission line. This is inconsistent with the Abell 2052 soft excess spectrum.
A strong soft component without emission lines may be caused by extended non-thermal emission. Since the cluster shows neither strong radio emission nor merger features, such a non-thermal emission could be due to a rather old population of non-thermal electrons (see e.g. Rephaeli et al. 2008 - Chapter 5, this volume).
5.4. Other clusters
The Sculptor supercluster was observed with Suzaku in 4 pointings with the fields partially overlapping and connecting 3 main clusters (Abell 2811 at z = 0.1086, Abell 2804 at z = 0.11245) and Abell 2801 at z = 0.11259). The data were analysed by Kelley et al. 2007. The combined spectrum after removing bright clusters and point sources shows an excess in the energy range 0.6-1 keV, with a temperature of about 0.8 keV. Since the observed volume is extremely large (~ 500 Mpc3), the uniform electron density implied is as low as 8 × 10-6 cm-3.
The Suzaku XIS data on Abell 1060 (z = 0.01140, Sato et al. 2007a), AWM7 (z = 0.01724, Sato et al. 2007b), and Fornax (Matsushita et al. 2007) have been analysed in some detail. Even though the spectrum and intensity of the foreground Galactic emission have fairly large ambiguities, the observed energy spectra for these clusters are generally well fit by thermal models at energies down to about 0.3 keV. For the central region of Fornax (r < 2'), a two temperature model with kT = 1.5 and 0.8 keV was preferred. However, excess emission which causes significant deviation from these standard spectral fits has not been detected from these clusters.
Fujita et al. (2007) observed the region between Abell 399 and Abell 401 with Suzaku. They found no evidence for oxygen emission from the WHIM in this region and obtained a strict upper limit of 4.1 × 10-5 cm-3 on its density.