Systematic X-ray Analysis of Radio Relic Clusters with
We perform a systematic X-ray analysis of six giant radio relics in four clusters of galaxies using the Suzaku satellite. The sample includes CIZA 2242.8-5301, Zwcl 2341.1-0000, the South-East part of Abell 3667 and previously published results of the North-West part of Abell 3667 and Abell 3376. Especially we first observed the narrow (50 kpc) relic of CIZA 2242.8-5301 by Suzaku satellite, which enable us to reduce the projection effect. We report X-ray detections of shocks at the position of the relics in CIZA2242.8-5301 and Abell 3667 SE. At the position of the two relics in ZWCL2341.1-0000, we do not detect shocks. From the spectroscopic temperature profiles across the relic, we find that the temperature profiles exhibit significant jumps across the relics for CIZA 2242.8-5301, Abell 3376, Abell 3667NW, and Abell 3667SE. We estimated the Mach number from the X-ray temperature or pressure profile using the Rankine-Hugoniot jump condition and compared it with the Mach number derived from the radio spectral index. The resulting Mach numbers (1.5-3) are almost consistent with each other, while the Mach number of CIZA2242 derived from the X-ray data tends to be lower than that of the radio observation. These results indicate that the giant radio relics in merging clusters are related to the shock structure, as suggested by previous studies of individual clusters.
reception date \Acceptedaccepted date \Publishedpublication date \SetRunningHead \KeyWordsgalaxies: clusters — galaxies: intergalactic medium — shock waves — X-rays: galaxies: clusters
The formation theory of galaxy clusters and groups is now well established: the high density peaks that exist within the primordial matter distribution later grow to groups and clusters by both accretion of surrounding matter and mergers of clumps. In the growth history of haloes, major mergers have a significant effect on the internal structure of clusters. X-ray observations have revealed a lot about the merging phenomena of haloes from the massive clusters scale (e.g. [Markevich et al. (2002)]) to galaxy groups (e.g. [Kawahara et al. (2011)]) mainly by studying X-ray morphology. Shock structures induced by mergers were clearly found in the Bullet cluster using Chandra (Markevich et al., 2002). In general, however, an identification of a shock structure in X-ray images is difficult, except in major mergers including 1E0657-56, A520, A2146 and A754 (Clowe et al. (2006); Markevitch et al. (2005); Russell et al. (2010); Macario et al. (2011)).
Recent progress in radio astronomy has revealed interesting arc structures in merging clusters, called radio relics (Ferrari et al., 2008). Radio relics are believed to be a tracer of shock structures through synchrotron radiation. Finoguenov et al. (2010) used XMM-Newton data of the well-known radio relic cluster Abell 3667 and found a sharp temperature decrement across the radio relic. Akamatsu et al. (2012) confirmed their result using data from the Suzaku satellite.
Here, we present a systematic X-ray analysis of 6 radio relics in 4 clusters observed using Suzaku, including a new observation of a strong radio relic cluster, CIZA 2242.8-5301 (hereafter CIZA 2242), which was recently discovered in a Giant Metrewave Radio Telescope (GMRT) observation (Swarup, 1991). New analyses of archival data of Zwcl 2341.1-0000 (hereafter Zwcl2341), Abell3667SE, and a compilation of previous results of Abell3667NE (Akamatsu et al., 2012) and Abell3376 (Akamatsu et al., 2011b) are also discussed.
We use km s Mpc, and , respectively. We use solar abundances as given by Anders & Grevesse (1989). Unless otherwise stated, the errors correspond to the 68% confidence level for a single parameter. The rest of this paper is organized as follows. In §2, we describe the target selection, data reduction, and the X-ray spectral analysis. In §3 we derive the temperature across the relic to estimate the Mach number. In §4 we summarize the results.
|redshift||physical length for||Temperature||Radio references|
|CIZA2242||0.192||190||8.4||33.4||van Weeren et al. (2010)|
|Abell3376||0.046||54||4.2||5.8||Bagchi et al. (2006), Kale et al. (2012)|
|Abell3667NW||0.056||66||6.0||4.7||Rötttgering et al. (1997)|
|Abell3667SE||0.056||66||6.0||4.7||Rötttgering et al. (1997)|
|Zwcl2341N||0.270||246||5.7||3.4||Bagchi et al. (2002); van Weeren et al. (2009)|
|Zwcl2341S||0.270||246||5.7||3.4||Bagchi et al. (2002); van Weeren et al. (2009)|
|:Akamatsu et al. (2012)|
|:Akamatsu et al. (2011b)|
|Name (ObsID)||(R.A., DEC )||Observation Start time||Exp time|
|CIZA2242 OFFSET(806002010)||(339.29, +52.67)||2011-07-30||52.4|
2 Observations & Data preparation
In Table 3, we summarize the information about the relics and clusters we analyze. Four relics (CIZA2242, Abell3667SE, Zwcl2341N, and Zwcl2341S) are newly analyzed in this paper and our previous analysis are used for two relics (Abell3667NW, A3376). We examine these clusters one by one.
CIZA2242 () was discovered in the Clusters in the Zone of Avoidance (CIZA) survey (Kocevski et al., 2007). van Weeren et al. (2010) reported a Mpc scale radio relic located in the Northern outskirts, at a distance of 1.2 Mpc from the cluster center. The radio relic is extremely narrow with a width of 55 kpc. The spectral index at the front of relic is -0.6 0.05, which corresponds to a Mach number of . Although this shock wave is the strongest of all the shocks in this sample, only X-ray data from ROSAT have been available so far. We obtained a 120 ksec Suzaku observation of CIZA2242 performed on September 28, 2011 with an additional 60 ksec offset observation.
A3376 () is considered to be experiencing a binary subcluster merger. Previous BeppoSAX/PDS observations showed a 2.7 detection of a hard X-ray signal Nevalainen et al. (2004). However, Suzaku HXD observations gave an upper limit, which did not exclude the BeppoSAX flux (Kawano et al., 2009). The detection of a hard X-ray signal is still controversial. Another striking feature of A3376 is a pair of Mpc-scale radio relics (Bagchi et al., 2006; Kale et al., 2012). Using Suzaku data, Akamatsu et al. (2011b) confirmed evidence for the presence of shock fronts across the west radio relic.
A3667 () is well known as a merging cluster with irregular X-ray morphology and has large two extended radio relics at the North-West (A3667NW) and the South-East (A3667SE) (Rötttgering et al., 1997). The overall spectral index of North-West relic gradually varies -1.1 to -1.5 toward the southeast direction. A3667 exhibits an elongated X-ray shape to the North-West direction with an average temperature of keV (Knopp et al., 1996; Briel et al., 2004). Recent XMM-Newton and Suzaku observations of the North-West radio relic revealed a significant jump in temperature and surface brightness across the relic (Finoguenov et al., 2010; Akamatsu et al., 2012). The observed ICM properties indicate the presence of a shock front with Mach number . The South-East radio relic has not yet been studied in X-rays.
Zwcl2341 () has two Mpc-scale structures of diffuse radio emission, which were discovered in 1.4 GHz New VLA All the Sky Survey (NVSS) (Bagchi et al., 2002) at the North (Zwcl2341N) and South (Zwcl2341S) of the clusters. These radio relics are located at a Northern (Zwcl2341N 860 kpc ) and Southern (Zwcl2341S 1100 kpc) direction from the cluster center (Bagchi et al., 2002). Recently, GMRT observations revealed that those radio emissions exhibit ”arc”-like shapes (van Weeren et al., 2009). The spectral indices of the relics are -0.490.18 for Zwcl2341N and -0.760.17 for Zwcl2341S corresponding to Mach numbers and , respectively. Previous Chandra and XMM-Newton observations (van Weeren et al., 2009) show that the X-ray emission extends over 3.3 Mpc in the North-South direction. Due to its low X-ray surface brightness, the X-ray information around these radio relics is still limited.
2.2 Data reduction
All SUZAKU observations of CIZA2242, CIZA2242 OFFSET, Abell3667SE and Zwcl2341 were performed with the XIS in either normal or clocking mode. The detailed information about these observations is summarized in Table 2. For all data, 3 out of the 4 CCD chips were available: XIS0, XIS1 and XIS3. The XIS1 is the back-illuminated chip with high sensitivity in the soft X-ray energy range.
We use HEAsoft version 6.11 and the calibration database (CALDB) dated 2011-06-30. We perform the event screening with the cut-off rigidity (COR) GV to increase the signal-to-noise ratio. For CIZA2242, we set the Earth rim ELEVATION to avoid contamination by scattered solar X-rays from the day Earth limb.
Furthermore, we apply additional processing for XIS1 to reduce NXB level, which
increased after changing the amount of charge injection (06/2010).
The detailed processing procedure is the same as the one described in
2.3 X-ray Spectrum Analysis
In this paper, we follow the same spectral fitting procedure as for the Suzaku observations of A3667 and A3376 (Akamatsu et al., 2012, 2011b). The basic method of the spectral fit is described in subsection 4.1 of Akamatsu et al. (2012).
Background and Foreground Emissions
An accurate measurement of the background components is crucial for the ICM study in the cluster outskirts. We consider three sky background components: the local hot bubble (LHB), the Milky way halo (MWH), and the cosmic X-ray background (CXB). We analysed the offset region for CIZA2242OFFSET (ID:806002010), which is one degree away from the main target. For the A3667SE observation, we used the background model reported in Akamatsu et al. (2012) Table3 and for Zwcl2341, we used the outer region where there is negligible ICM emission.
The non X-ray background (NXB) subtracted spectra were fitted by the summation of the LHB: (0.1 keV), the MWH (0.3 keV) and the CXB as . We fix the redshift and abundance of all apec components to zero and unity, respectively. The NXB component was estimated from the dark Earth database by the FTOOLS (Tawa et al., 2008) and was subtracted from the data before the spectral fit. We used the XIS response assuming uniform brightness on the sky by xissimarfgen (Ishisaki et al., 2007). The best fit parameters and spectrum of CIZA2242OFFSET are shown in Table 3 and Fig. 3, respectively.
|9.6 for in the unit of photons keV at 1 keV.|
|The referred CXB intensity normalization in Kushino et al. (2002) is|
|where is the angular diameter distance to the source.|
|*:Normalization of the apec component scaled with a factor 1/400.|
We extracted spectra from regions defined by their distances from the center () (Fig 1a ) for CIZA2242, centered on () (Fig 1b ) for A3667SE, centered on () for Zwcl2341N, centered on () (Fig 1d ) for Zwcl2341S, respectively. For CIZA2242, the center was determined by visually fitting a circle to the radio relics. For other clusters, the center was determined from their X-ray surface brightness peaks.
The response matrix files (RMFs) and ancillary response files (ARFs) were generated using xisrmfgen and xissimarfgen (Ishisaki et al., 2007). Since the Galactic and CXB components can be regarded as being spatially uniform in the XIS field of view, we use an uniform ARF for these components. For the ICM emission, we need to put the flux distribution into xissimarfgen. For A3667 and A3376, we adopted the -model distribution with =0.54, =\timeform2.97’ and , , respectively. For CIZA2242 and Zwcl2341, we used the Suzaku XIS image (0.5-8.0 keV).
The spectral fitting of each region was performed as described below. We modeled the NXB subtracted spectrum as the sum of the ICM emission and each background component discussed in Sec 2.3.1. Intensity, temperature, photon indices of CXB and galactic components were all fixed at the value determined in offset observation. The interstellar absorption was kept fixed using the twenty-one centimeter line measurement of the hydrogen column, for CIZA2242, A3667SE and Zwcl2341, respectively (Dickey & Lockman, 1990). We simultaneously fitted both the BI and FI spectra with XSPEC ver12.7. For the inner regions of the radio relic, the free parameters were temperature , normalization norm and metal abundance of the ICM component. Otherwise, we fixed the ICM metal abundance to the typical value in cluster outskirts, (e.g. Fujita et al. (2008)). The relative normalization between the two sensors was a free parameter in this fit to compensate for cross-calibration errors.
3.1 ICM properties of individual cluster
We first perform the spectral fitting within the radio relic ( 1.2 Mpc=): the temperature and the metal abundance are keV and , respectively. The unabsorbed flux and luminosity within the radio relic in the 0.510 keV band are and , respectively.
Then, we performed the spectral fitting for each annulus examined in Sec 2.3.2. Figure 2 displays spectra of six annuli. Figure 4(a) shows the radial profile of the ICM temperature in CIZA2242. The temperature within (1.25 Mpc) from the cluster center shows a fairly constant value of keV. As mentioned in section1, van Weeren et al. (2010) reported the sharp radio relic at that radius, which is indicated by vertical dashed line in Panel (a). We found that the temperature profile exhibits a significant jump from 8.3 keV to 2.1 keV across the radio relic. Because the ICM emission is lower than the background level in the outer bins (), we consider the variability of the CXB as a systematic error. The fluctuation of CXB intensity was estimated based on our previous work (Akamatsu et al., 2011a). The estimated fluctuations span 15-25 %. The resultant parameters after taking into account the systematic error are shown blue dotted lines in figure 4 (a). The temperature jump is consistent with the prediction of numerical simulation by van Weeren et al. (2011). For a reference, we plotted the ”universal” temperature profile expected from the scaled temperature profiles (Burns et al., 2010).
As shown in Fig. 5(a), we could not see the surface brightness jump across the relic directly in the profile. However, in Appendix we show that this is caused by the limited Suzaku spatial resolution, and the observed profile is in fact consistent with the expected density jump
Previous studies of A3376 found that the global mean temperature is 4.0 keV and showed the presence of a pair of Mpc-scale radio relics consistent with 1.4 GHz VLA NVSS observations (Bagchi et al., 2006). A detailed study of the A3376 radio relic has already been published by Akamatsu et al. (2011b). We refer to that paper for the images and X-ray information and we will include A3376 in the discussion.
Suzaku performed 3 pointing observations of Abell 3667 along the North-West merger axis in May 2006. Figure. 1(c) displays the NXB subtracted X-ray image of A3667. A detailed study of A3667NW radio relic has already been published by Akamatsu et al. (2012). We refer to that paper for the images and X-ray information and we will include A3667NW in the discussion.
Figure. 4(d) indicates the radial profile of the ICM temperature in the South-East direction of A3667, respectively. The innermost 2 bins of the cluster are identical to that of A3667NW. The temperature within the SE radio relic ( 1.3 Mpc) from the cluster center shows a fairly constant value of keV, which is consistent with the NW direction (Akamatsu et al., 2012). Our temperature profile exhibits a jump across the SE radio relic region. Although we could not see the surface brightness jump (Fig. 5d), the pressure profiles (Fig. 6d) exhibit a slight drop due to the temperature jump.
We first extract the spectrum within 3’ ( 740 kpc) and fix the redshift value to 0.270. The spectrum was well fitted with the above model ( for 473 degrees of freedom). We find a temperature and metal abundance of keV and , respectively. The unabsorbed flux and luminosity in the 0.510 keV band are and , respectively. The derived temperature is consistent with previous Chandra observations ( 5 keV) (van Weeren et al., 2009).
The radial profiles of the ICM temperature are shown in Fig. 4(e) and (f) for Zwcl2341N and Zwcl2341S, respectively. The temperature drops from 4keV to 1 keV toward the Northern direction or from 4 keV to 2 keV toward the Southern direction. Figure 5(e,f) and Figure 6(e,f) show the surface brightness and the pressure profiles of Zwcl2341N and Zwcl2341S. We could not see any jump in our wide radial bins for temperature, surface brightness, and pressure profiles for this cluster.
|: The first and the second errors are statistical and systematic (CXB fluctuations).|
3.2 The scaled temperature profiles
We normalized the radial temperature profile of each cluster by the flux-weighted average temperature. The resulting scaled temperature profiles are shown in Fig 7. We find that the temperature profiles of four relics, including the previous two results, exhibit significant discontinuities across the radio relic. These discontinuities can be associated with the shock front found in the radio relics. For the other two samples (Zwcl2341N,S), we could not see any jump in temperature profiles. The angular resolution of Suzaku was not enough to check the presence of a discontinuity. Therefore, we focus on these four examples of temperature discontinuity.
Another remarkable point of the scaled temperature profile is its flatness within the radio relic. Recent X-ray results show that the ICM in relaxed clusters has the ”universal” decline temperature profile up to virial radius (Markevitch et al., 1998; De Grandi & Molendi, 2002; Vikhlinin et al., 2005; Pratt et al., 2007; Akamatsu et al., 2011a). The flatness of temperature within the radio relic might need a heating mechanism, such as shock heating.
As mentioned above, the temperature of relaxed clusters tends to have the ”universal” profile. It is natural that the flat temperature profiles of merging cluster will settle to the universal profile in less than a Hubble time. In our samples, the locations of the radio relic are far from cluster center, typically Mpc (). Due to the low density () in the outskirts of clusters, the cooling time in those regions exceeds the Hubble time. We might consider other processes to achieve the universal profile.
3.3 Mach number at Discontinuity
We estimate the Mach numbers using temperature jumps across the relics. The Mach number can be obtained by applying the Rankine-Hugoniot jump condition, assuming the ratio of specific heats as , as
are the pre-shock and post-shock temperatures and is the estimated Mach number. The post and pre quantities we use are marked by red and blue horizontal bars in Figure 4.
We compare our results with the Mach number derived from previous radio observations. Based on Diffusive Shock Acceleration (DSA) theory, we estimate the Mach number based on the radio spectrum, as
where is radio spectral index and is the Mach number from the radio observation. For strong shocks, the radio spectral index will be flat about -0.5. We used radio spectral indices from the radio observations summarized in Table 1. The resulting Mach numbers are shown in Table 4. Figure 8 displays the Mach number from the X-ray temperature and radio observations. As shown in this figure, we find that the two different methods give consistent results within the current observational noise.
If we assume in the case of CIZA2242 that the mach number derived from the radio observation is correct, the expected post shock temperature would reach k keV. Although our results are well reproduced by a 1 temperature model, we also fit 2 temperature models to the spectrum extracted from the \timeform5.0’ annulus, which is located in the region where the shock just passed. The metal abundances of the two apec components were tied to be the same, and only the temperature and normalization were set free. We then obtained a statistically acceptable fit with / d.o.f. = 228/244 compared with the single temperature case of 231/246. The rather low statics of the spectrum hampered us to distinguish between the and models. The hot component temperature was derived as keV, and the cool one as keV. The intensity of the cool component is about a order lower than hot one. Based on those analyses, there were no signatures of the presence of high-temperature components.
From simulations of the structure formation of the Universe (Enßlin & Brüggen, 2002; Vazza et al., 2009), large accretion shocks and low merger shocks around clusters of galaxies are expected. To understand the physics of cluster evolution, cluster merger events are important process. Suzaku XIS is best suited instrument for measuring the ICM emission beyond the radio relics thanks to its low and stable detector background. In this paper, we analyse six radio relics of four clusters with Suzaku XIS and measured their temperature profiles. We found clear temperature jumps across four relics and derived the Mach number from the pre- and post temperature. The derived Mach number is almost consistent with that from the radio observation. These results strongly suggest that a shock structure exists at the radio relic.
We are grateful to Takaya Ohashi and Jelle de Plaaa for insightful discussions and comments and Ruta Kale for kindly providing the radio index data of A3376. The authors thank the anonymous referee for insightful comments and suggestions which improved the content of the paper. The authors thank the Suzaku team members for their support of the Suzaku project. H. A. and H.K. are supported by a Grant-in-Aid for Japan Society for the Promotion of Science (JSPS) Fellows (221582 and 225467).
Appendix A Surface brightness jump in CIZA2242
Because of the non-detection of surface brightness jump in CIZA2242, we performed an additional check. Based on previous works (e.g., Markevich et al. (2002)), we modeled the electron density profile switching at radio relic () from the beta model to a power law. In this model, we assume that two cases of the jump amplitude, corresponding to the Mach number (our lower limit including the systematic uncertainty, see Table 4) and M=4.5 (the radio value). Assuming spherical symmetry, we calculated the square of density as surface brightness and convolved it with the Suzaku PSF. The profile was binned with the same size as the observed surface brightness. Because the jump in the surface brightness is important, we selected the profile across the radio relic (). Finally, we normalized the profile to the observed value at the post shock region (). The resulting profiles are shown in Fig. 9b (Blue and Black). The observed brightness profile, shown by red crosses, is consistent with the interval of M obtained from our temperature measurements.
For the CIZA2242 case, the PSF of Suzaku corresponds to 380 kpc, which is probably much larger than the length of shock. Hence the surface brightness jump should be significantly diluted by other area in the bin. We need more observations with a higher angular resolution to confirm the shock structure in the surface brightness profile.
(a)Modeled density profile
(b)Modeled surface brightness profile
- thanks: Last update: March 7, 2018
- affiliation: SRON Netherlands Institute for Space Research, Sorbonnelaan 2, 3584 CA Utrecht, The Netherlands
Department of Physics, Tokyo Metropolitan University,
1-1 Minami-Osawa, Hachioji, Tokyo 192-0397
- Akahori, T., & Yoshikawa, K. 2010, \pasj, 62, 335
- Akamatsu, H., Hoshino, A., Ishisaki, Y., et al. 2011, \pasj, 63, 1019 (2011a: A2142)
- Akamatsu, H., de Plaa, J., Kaastra, J., et al. 2012, \pasj, 64, 49
- Akamatsu, H., Takizawa, M., Nakazawa, K., et al. 2011, arXiv:1112.5955 (2011c: A3376)
- Anders, E., & Grevesse, N. 1989, \gca, 53, 197
- Bagchi, J., Enßlin, T. A., Miniati, F., et al. 2002, , 7, 249
- Bagchi, J., Durret, F., Neto, G. B. L., & Paul, S. 2006, Science, 314, 791
- Briel, U. G., Finoguenov, A., & Henry, J. P. 2004, \aap, 426, 1
- Borgani, S., et al. 2004, \mnras, 348, 1078
- Burns, J. O., Skillman, S. W., & O’Shea, B. W. 2010, \apj, 721, 1105
- Clowe, D., Bradač, M., Gonzalez, A. H., Markevitch, M., Randall, S. W., Jones, C., & Zaritsky, D. 2006, \apjl, 648, L109
- De Grandi, S., & Molendi, S. 2002, \apj, 567, 163
- Dickey, J. M., & Lockman, F. J. 1990, \araa, 28, 215
- Enßlin, T. A., & Brüggen, M. 2002, \mnras, 331, 1011
- Ettori, S., et al. 2004, \mnras, 354, 111
- Ferrari, C., Govoni, F., Schindler, S., Bykov, A. M., & Rephaeli, Y. 2008, \ssr, 134, 93
- Finoguenov, A., Sarazin, C. L., Nakazawa, K., Wik, D. R., & Clarke, T. E. 2010, apj, 715, 1143
- Fujita, Y., et al. 2008, \pasj, 60, 1133
- Ishisaki, Y., et al. 2007, \pasj, 59, 113
- Kale, R., Dwarakanath, K. S., Bagchi, J., & Paul, S. 2012, arXiv:1206.3389
- Kawahara, H., Yoshitake, H., Nishimichi, T., and Sousbie, T.. 2011, \apjl, 727, L38
- Kawano, N., et al. 2009, \pasj, 61, 377
- Knopp, G. P., Henry, J. P., & Briel, U. G. 1996, \apj, 472, 125
- Kocevski, D. D., Ebeling, H., Mullis, C. R., & Tully, R. B. 2007, \apj, 662, 224
- Kushino, A., Ishisaki, Y., Morita, U., Yamasaki, N. Y., Ishida, M., Ohashi, T., & Ueda, Y. 2002, \pasj, 54, 327
- Macario, G., Markevitch, M., Giacintucci, S., et al. 2011, \apj, 728, 82
- Markevitch, M., Forman, W. R., Sarazin, C. L., & Vikhlinin, A. 1998, \apj, 503, 77
- M. Markevitch, A. H. Gonzalez, L. David, A. Vikhlinin, S. Murray, W. Forman, C. Jones, and W. Tucker. 2002, \apjl, 567,L27–L31
- Markevitch, M., Govoni, F., Brunetti, G., & Jerius, D. 2005, \apj, 627, 733
- Markevitch, M. 2006, The X-ray Universe 2005, 604, 723
- Mitsuda, K., et al. 2010, \procspie, 7732,
- Nevalainen, J., Oosterbroek, T., Bonamente, M., & Colafrancesco, S. 2004, \apj, 608, 166
- Pratt, G. W., Böhringer, H., Croston, J. H., Arnaud, M., Borgani, S., Finoguenov, A., & Temple, R. F. 2007, \aap, 461, 71
- Röttgering, H. J. A., Wieringa, M. H., Hunstead, R. W., & Ekers, R. D. 1997, \mnras, 290, 577
- Rybicki, G. B., & Lightman, A. P. 1979, New York, Wiley-Interscience, 1979. 393 p.,
- Rudd,D.H., & Nagai,D., 2009, \apj, 701, L16
- Russell, H. R., Sanders, J. S., Fabian, A. C., Baum, S. A., Donahue, M., Edge, A. C., McNamara, B. R., & O’Dea, C. P. 2010, \mnras, 406, 1721
- Russell, H. R., McNamara, B. R., Sanders, J. S., et al. 2012, arXiv:1202.5320
- Spitzer,L.Jr., 1956, Physics of Fully Ionized Gases, New York: Interscience
- Swarup, G. 1991, IAU Colloq. 131: Radio Interferometry. Theory, Techniques, and Applications, 19, 376
- Takizawa,M., 1998, \apj, 509, 579
- Takizawa, M. 1999, \apj, 520, 514
- Takizawa, M. 2005, \apj, 629, 791
- Tawa, N., et al. 2008, \pasj, 60, 11
- van Weeren, R. J., Röttgering, H. J. A., Bagchi, J., et al. 2009, \aap, 506, 1083
- van Weeren, R. J., Röttgering, H. J. A., Brüggen, M., & Hoeft, M. 2010, Science, 330, 347
- van Weeren, R. J., Brüggen, M., Röttgering, H. J. A., & Hoeft, M. 2011, \mnras, 1445
- Vazza, F., Brunetti, G., & Gheller, C. 2009, \mnras, 395, 1333
- Vikhlinin, A., Markevitch, M., Murray, S. S., Jones, C., Forman, W., & Van Speybroeck, L. 2005, \apj, 628, 655