当前位置:首页 >> >>

Hard X-ray emission from the galaxy cluster A3667

Hard X-ray emission from the galaxy cluster A3667
Roberto Fusco-Femiano Istituto di Astro?sica Spaziale, C.N.R., via del Fosso del Cavaliere, I-00133 Roma, Italy -

arXiv:astro-ph/0105049v1 3 May 2001

dario@saturn.ias.rm.cnr.it Daniele Dal Fiume? TESRE, C.N.R., via Gobetti 101, I-40129 Bologna, Italy Mauro Orlandini TESRE, C.N.R., via Gobetti 101, I-40129 Bologna, Italy Gianfranco Brunetti Dipartimento di Astronomia, Univ. Bologna, via Ranzani 1, I-40127 Bologna, Italy Luigina Feretti Istituto di Radioastronomia, C.N.R., via Gobetti 101, I-40129 Bologna, Italy Gabriele Giovannini Istituto di Radioastronomia, C.N.R., via Gobetti 101, I-40129 Bologna, Italy Dip. di Fisica, Univ. Bologna, via B. Pichat 6/2, I-40127 Bologna, Italy

We honour our late colleague for his invaluable contribution to the search for

nonthermal emission in clusters of galaxies.





We report the results of a long BeppoSAX observation of Abell 3667, one of the most spectacular galaxy clusters in the southern sky. A clear detection of hard X-ray radiation up to ?35 keV is reported, while a hard excess above the thermal gas emission is present at a marginal level that should be considered as an upper limit to the presence of nonthermal X-ray radiation. The strong hard excesses reported by BeppoSAX in Coma and A2256 and the only marginal detection of nonthermal emission in A3667 can be explained in the framework of the inverse Compton model. We argue that the nonthermal X-ray detections in the PDS energy range are related to the radio index structure of halos and relics present in the observed clusters of galaxies.

Subject headings: cosmic microwave background — galaxies: clusters: individual (A3667) — magnetic ?elds — radiation mechanisms: non-thermal — X-rays: galaxies

–3– 1. Introduction

Abell 3667 is one of the most intriguing clusters of galaxies. It contains one of the largest radio sources in the southern sky with a total extent of ? 30′ which corresponds to ? 2.6h?1 Mpc. This di?use radio emission is located to the north-west, well outside 50 the central core of the cluster. A similar but weaker radio region is present also to the south-east, yielding a very peculiar structure not observed in any other cluster (Robertson 1991; Rottgering et al. 1997). The Mpc-scale radio relics may be originated by the ongoing merger visible in the optical (Sodre’ et al. 1992), in X-ray (Knopp, Henry & Briel 1996) and in the weak lensing mass map (Jo?re et al. 2000). In the optical, A3667 shows strong galaxy concentrations around the two brightest D galaxies (Proust et al. 1988; Sodre’ et al. 1992). The main optical component is coincident with the peak of the X-ray emission. The ASCA observation reports an average gas temperature of 7.0±0.6 keV with a constant radially averaged pro?le up to ? 22′. The ASCA temperature map shows that the hottest region is in between the two groups of galaxies con?rming the merger scenario (Markevitch, Sarazin & Vikhlinin 1999). Recently, a Chandra observation reported a sharp X-ray brightness edge near the cluster core (Vikhlinin, Markevitch & Murray 2000) due to a large cool gas cloud moving through the hot intracluster medium (ICM). Major cluster mergers can produce large-scale shocks and turbulence in the ICM establishing conditions for in-situ particle reacceleration and magnetic ?eld ampli?cation (Tribble 1993; Longair 1994, and references therein). The nonthermal electron population produces emission in various energy bands. Extended radio sources are the most evident signatures of nonthermal processes in clusters of galaxies. More recently, the detections in the Coma cluster (Fusco-Femiano et al. 1999; Rephaeli, Blanco & Gruber 1999) and A2256 (Fusco-Femiano et al. 2000) of a hard X-ray excess with respect to the thermal emission represent a further evidence of nonthermal processes in some clusters of galaxies. This

–4– allows to derive additional information on the physical conditions of the ICM environment, which cannot be obtained by studying the thermal plasma emission only. Both the clusters where nonthermal X-ray emission has been detected show extended radio halos in the center. Moreover a relic region is present in A2256, at ? 8′ from the center, more extended and brighter than the halo source (Bridle et al. 1979). A marginal evidence of nonthermal radiation is reported in the external regions of the MECS detector for A2199 (Kaastra et al. 1999), a cluster which does not show any extended radio region. So, the most direct explanation for the detected hard X-ray excesses is that they are due to inverse Compton (IC) scattering of cosmic microwave background (CMB) photons by the relativistic electrons responsible for the radio emission. Alternative interpretations to the IC model for the nonthermal radiation detected in the Coma cluster have been proposed. Blasi & Colafrancesco (1999) suggest a secondary electron production. However, this model implies a γ-ray ?ux considerably larger than the EGRET upper limit unless the hard X-ray excess and the radio halo emission in Coma are not due to the same population of electrons. A di?erent mechanism is given by nonthermal bremsstrahlung from suprathermal electrons formed through the current acceleration of the thermal gas (Ensslin et al. 1999; Dogiel 2000; Sarazin & Kempner 2000; Blasi 2000). At present, due to the low e?ciency of the proposed acceleration processes and of the bremsstrahlung mechanism, these models would require an unrealistically high energy input as recently pointed out by Petrosian (2001). Besides, this interpretation seems to be in con?ict with the BeppoSAX detection of A2256, since the derived ?at power-law spectrum of the electrons gives a negligible nonthermal bremsstrahlung contribution to the PDS ?ux (Fusco-Femiano et al. 2000). These alternative models are motivated by the discrepancy between the value for the intracluster magnetic ?eld derived by the BeppoSAX observation of the Coma cluster (BXR ? 0.16?G, Fusco-Femiano et al. 1999) and the value derived from Faraday rotation (FR) of polarized radiation toward the radio galaxy NGC4869 (BF R ? 6?G,

–5– Feretti et al. 1995). However, this discrepancy could be resolved considering that Feretti et al. (1995) inferred also the existence of a weaker and larger scale magnetic ?eld in the range 0.1-0.2 h50 ?G consistent with the value measured by BeppoSAX , and therefore the component of ? 6?G could be local. Goldshmidt & Rephaeli (1993) suggested that the discrepancy between BXR and BF R could be alleviated taking into consideration the expected spatial pro?les of the magnetic ?eld and relativistic electrons. More recently, it has been shown that IC models which include the e?ects of more realistic electron spectra, combined with the expected spatial pro?les of the magnetic ?eld, and anysotropies in the pitch angle distribution of the electrons allow higher values of the intracluster magnetic ?eld in better agreement with the Faraday rotation measurements (Brunetti et al. 2001; Petrosian 2001). In this Letter, we present the results of a long observation of A3667, exploiting the unique capabilities of the PDS (Frontera et al. 1997) onboard BeppoSAX , to search for hard X-ray radiation (HXR) emission. Besides, we discuss a possible explanation for the marginal detection of the nonthermal hard excess with respect to the thermal emission reported by this observation. Throughout the Letter we assume a Hubble constant of Ho = 50 km s?1 Mpc?1 h50 and q0 = 1/2, so that an angular distance of 1′ corresponds to 87 kpc (zA3667 = 0.055, Sodre’ et al. 1992). Quoted con?dence intervals are at 90% level, if not otherwise speci?ed.


PDS Data Reduction and Results 20h 11m 30.s ;

The pointing coordinates of BeppoSAX are at J(2000): α :

δ : ? 56? 40′ 0′′ .0, in proximity of the secondary galaxy concentration and the PDS FOV (1.3? ) includes only the radio region in the north of the cluster. The total e?ective exposure

–6– time for the PDS was ? 1.13 × 105 sec in the two observations of May 1998 and October 1999. Since the source is rather faint in the PDS band (approximately 0.44 mCrab in 15-40 keV) a careful check of the background subtraction must be performed. The background sampling was performed using the default rocking law of the two PDS collimators that samples ON, +OFF, ON, -OFF ?elds for each collimator with a dwell time of 96 sec (Frontera et al. 1997). When one collimator is pointing ON source, the other collimator is pointing toward one of the two OFF positions. We used the standard procedure to obtain PDS spectra (Dal Fiume et al. 1997); this procedure consists of extracting one accumulated spectrum for each unit for each collimator position. We then checked the two independently accumulated background spectra in the two di?erent +/-OFF sky directions, o?set by 210′ with respect to the on-axis pointing direction. The comparison between the two accumulated backgrounds ([+OFF] vs. [-OFF]) does not show di?erence in the two pointings. The background level of the PDS is the lowest and more stable obtained so far with the high-energy instruments on board satellites thanks to its equatorial orbit. No modeling of the time variation of the background is required. The correctness of the PDS background subtraction has been checked by verifying that the counts ?uctuate at about zero ?ux as the signal falls below detectability. This happens at energies greater than ?35 keV. Figure 1 shows a clear detection of hard X-ray emission up to ?35 keV, at a con?dence level of ? 10σ. The observed count rate was 0.134±0.013 cts/s. The ?t to the PDS data with a thermal component (continuous line) at the ?xed average gas temperature of 7 keV derived from the ASCA observation (see Introduction) indicates a marginal presence of a hard excess at a con?dence level of ? 2.6σ. If we introduce a second nonthermal component, modeled as a power law, we obtain an insigni?cant improvement with respect to

–7– the previous model, according to the F-test. The low statistics do not allow us to perform a ?t with a thermal component at the ?xed average cluster gas temperature and a second thermal component. The only possibile ?t is with a thermal bremsstrahlung model that gives a best ?t temperature of ? 11 keV with a large con?dence interval (7.7-16.8 keV, 90%). The average cluster gas temperature is only slightly out of this interval. Therefore, the ? 2.6σ excess should be considered as an upper limit to the presence of nonthermal HXR radiation. The analysis of the two observations with e?ective exposure times of ?44 ks (1998 May) and ?69 ks (1999 October) for the PDS does not show signi?cant ?ux variations.



The long observation by BeppoSAX of Abell 3667 reports a robust evidence for hard X-ray emission in the PDS energy range 15-35 keV and an upper limit for a hard nonthermal excess with respect to the thermal emission at the average gas temperature of 7 keV. As pointed out in the Introduction, the most striking feature of Abell 3667 is the di?use arc-shaped radio region located to the NW of the X-ray core with a total extent of ? 30′ (? 2.6 Mpc), one of the largest radio sources known in the southern sky. The total ?ux density is 5.5±0.5 Jy at 843 MHz. Also the spectral index structure is very interesting: a region with a ?at spectrum (αr ? 0.5) in the NW rim of the source with a considerable steepening to αr ?1.5 toward the SW (Rottgering et al. 1997). The overall radio spectral index of ?1.1 is consistent with the ?ux density measurements (Mills et al. 1961; Bolton, Gardner & Mackey 1964; Robertson 1991). Fitting the PDS data with a thermal component, at the temperature of 7 keV, and a power law component, with index 2.1 (1+αr ), we derive a nonthermal IC ?ux upper limit of ? 4.1 × 10?12 erg cm?2 s?1 in the 15-35 keV energy. Extrapolating this ?ux in the energy range 20-80 keV we obtain

–8– FX ? 6.4 × 10?12 erg cm?2 s?1 that is a factor ?3.4 and ?2 lower than the nonthermal ?uxes detected in Coma and A2256, respectively. In the IC interpretation this ?ux upper limit, combined with the radio synchrotron emission, determines a lower limit to the volume-averaged intracluster magnetic ?eld of 0.41?G. Given the presence of such a large radio region in the NW of A3667, a robust detection of a nonthermal X-ray component might be expected instead of the upper limit reported by BeppoSAX . This result cannot be attributed to the shift (? 17′ ) of the PDS pointing with respect to the centroid of the radio relic (α : 20h 10m 30s .0; δ : ?56? 25′ 0.′′ 0). Instead, one possibile explanation may be related to the radio spectral index structure of the NW relic. As indicated by Roettiger, Burns & Stone (1999) the sharp edge of the radio source is the site of particle acceleration, while the progressive index steepening with the increase of the distance from the shock would indicate particle aging because of radiative losses. In the narrow shocked region, where particle reacceleration is at work, the magnetic ?eld is expected to be ampli?ed by adiabatic compression, with the consequence that the synchrotron emission is enhanced, thus giving a limited number of electrons the ability to produce IC X-rays. In the postshock region of the relic, the electrons su?er strong radiative losses with no reacceleration, considering also that the relic is well outside the cluster core. Therefore, their energy spectrum develops a high energy cuto? at γ < 104 , and the electron energy is not su?ciently high to emit IC radiation in the hard X-ray band. Synchrotron emission is detected from the postshocked region, because the magnetic ?eld is still strong due to its likely long relax time. The Coma cluster, where nonthermal HXR emission is present at a signi?cant level (Fusco-Femiano et al. 1999), shows a quite di?erent radio index structure (Giovannini et al. 1993; Deiss et al. 1997). The cluster exhibits a central plateau (R? 10′ ) with radio spectral index ?0.7 (in the core, it appears to be lower; Deiss et al. 1997) and a progressive

–9– spectral steepening with the increasing radius in the external regions of the radio halo. Moreover, the total extent of the radio halo (R? 80′ ) and the lack of a clear shocked region are not compatible with the scenario of the pure spectral aging of the emitting electrons. Thus, in situ reacceleration processes are required, probably because of turbulence related to recent mergers (Colless & Dunn 1996, Donnelly et al. 1999; Arnaud et al. 2001) with a possible additional contribution from the gas motion originating from the massive galaxies orbiting in the cluster core (Deiss & Just 1996). Brunetti et al. (2001) have recently shown that the radio observational properties of the Coma halo and the nonthermal HXR emission can be accounted for by a population of reaccelerated relativistic electrons with energy > break γb ? 104 emitting in a magnetic ?eld smoothly decreasing from the center toward the periphery. Similar considerations can be applied to Abell 2256, that is the second cluster that shows a clear evidence of a hard excess above the thermal intracluster emission (Fusco-Femiano et al. 2000). Abell 2256 exhibits a large di?use radio region (1.0×0.3 Mpc) in the north, at a distance of ? 8′ from the cluster center (relic), with a rather uniform and ?at spectral index of 0.8±0.1 between 610 and 1415 MHz. A fainter extended emission (halo) permeates the cluster center with a steeper radio spectral index of ?1.8 (Bridle & Fomalont 1976; Bridle et al. 1979; Rottgering et al. 1994; Rengelink et al. 1997). The low and uniform value of αr in the cluster relic indicates a broad reacceleration region, probably the result of an ongoing merger event (Sun et al. 2001). Moreover, the presence of the central radio halo would favour the hypothesis that in situ reacceleration processes are active in the cluster volume. As discussed by Fusco-Femiano et al. (2000), in the framework of the IC model, the probable source of the non-thermal HXR emission is the large relic if the associated magnetic ?eld is of the order of ? 0.1 ?G. An additional contribution could be provided by the halo electrons if the radio index structure is similar to that of the Coma cluster.

– 10 – 4. Conclusions

The positive detections of nonthermal HXR radiation in the Coma cluster and A2256 and the upper limit reported in A3667 by BeppoSAX are explained in the framework of the IC model. In particular, these results appear to be related to the radio spectral index structure of the radio halos or relics present in these clusters. The essential requirement to detect nonthermal X-ray emission in the PDS energy range is the presence of large regions of reaccelerated electrons, with γ ? 104 , due to the balance between radiative losses and reacceleration gains in turbulence generated by recent merger events.



We thank the referee for valuable comments and suggestions.

– 11 – REFERENCES Arnaud, M. et al. 2001, A&A, in press Blasi, P., & Colafrancesco, S. 1999, APh, 12, 169 Blasi, P. 2000, ApJ, 532, L9 Bolton, J.G., Gardner, F.F., & Mackey, M.B. 1964, Aust. J. Phys., 17, 340 Bridle, A., & Fomalont, E. 1976, A&A, 52, 107 Bridle, A., & Fomalont, E., Miley, G., & Valentijn, E. 1979, A&A, 80, 201 Brunetti, G., Setti, G., Feretti, L., & Giovannini, G., 2001, MNRAS, 320, 365 Colless, M., & Dunn, A. 1996, ApJ, 458, 435 Dal Fiume, D., Frontera, F., Nicastro, L., Orlandini, M., Palazzi, E., Costa, E., Feroci, M., & Zavattini, G. 1997, Proc. of the Fifth Workshop ”Data Analysis in Astronomy”, eds.: V. Di Gesu’ et al. , pag. 111 Deiss, B.M., Reich, W., Lesch, H., & Wielebinski, R. 1997, A&A 321, 55 Deiss, B.M., & Just, A. 1996, A&A 305, 407 Dogiel, V.A. 2000, A&A, 357, 66 Donnelly, R.H., Markevitch, M., Forman, W., Jones, C., Churazov, E., & Gilfanov, M. 1999, ApJ, 513, 690 Ensslin, T.A., Lieu, R., & Biermann, P.L. 1999, A&A, 344, 409 Feretti, L., Dallacasa, D., Giovannini, G., & Tagliani, A. 1995, A&A, 302, 680 Frontera, F., Costa, E., Dal Fiume, D., Feroci, M., Nicastro, L., Orlandini, M., Palazzi, E., & Zavattini, G. 1997, A&AS, 122, 357 Fusco-Femiano, R., Dal Fiume, D., Feretti, L., Giovannini, G., Grandi, P., Matt, G., Molendi, S., & Santangelo, A. 1999, ApJ, 513, L21

– 12 – Fusco-Femiano, R., Dal Fiume, D., De Grandi, S., Feretti, L., Giovannini, G., Grandi, P., Malizia, A., Matt, G., Molendi, S. 2000, ApJ, 534, L7 Giovannini, G., Feretti, L., Venturi, T., Kim, K.T., & Kronberg, P.P. 1993, ApJ, 406, 399 Goldshmidt, O., & Rephaeli, Y. 1993, ApJ, 411, 518 Jo?re, M. et al. 2000, ApJ, 534, L131 Kaastra, J.S., Lieu, R., Mittaz, J.P.D., Bleeker, J.A.M., Mewe, R., Colafrancesco, S., & Lockman, F.J. 1999, ApJ, 519, L119 Knopp, G.P., Henry, J.P., & Briel, U.G. 1996, ApJ, 472, 125 Longair, M.S. 1994, High Energy Astrophysics, Cambridge, UK : Cambridge Univ. Press Markevitch, M., Forman, W.R., Sarazin, C.L. & Vikhlinin, A. 1998, ApJ, 503, 77 Markevitch, M., Sarazin, C.L. & Vikhlinin, A. 1999, ApJ, 521, 526 Mills, B.Y., Slee, O.B., & Hills, E.R. 1961, Aust. J. Phys., 14, 497 Petrosian, V. 2001, astro-ph/0101145 Proust, D., Mazure, A., Sodre’, L., Capelato, H.V., & Lund, G. 1988, A&AS, 72, 415 Rengelink, R.B. et al. 1997, A&AS, 124, 259 Rephaeli, Y., Gruber, D.E., & Blanco, P. 1999, ApJ, 511, L21 Robertson , J.G. 1991, Aust. J. Phys., 44, 729 Roettiger, K., Burns, J.O., & Stone, J.M. 1999, ApJ, 518, 603 R¨ttgering, H., Snellen, I., Miley, G., de Jong, J.P., Hanish, R.J., & Perley, R. 1994, ApJ, o 436, 654 R¨ttgering, H.J.A., Wieringa, M.H., Hunstead, R.W., & Ekers, R.D. 1997, MNRAS, 290, o 577 Sarazin, C.L., & Kempner, J.C. 2000, ApJ, 533, 73

– 13 – Sodre’, L., Capelato, H.V., Steiner, J., Proust, D., & Mazure, A. 1992, MNRAS, 259, 233 Sun, M., Murray, S.S., Markevitch, M. & Vikhlinin, A. 2001, astro-ph/0103103 Tribble, P.C. 1993, MNRAS, 261, 57 Vikhlinin, A., Markevitch, M., & Murray, S.S. 2000, astro-ph/0008496

A This manuscript was prepared with the AAS L TEX macros v4.0.

– 14 – Fig. 1.— PDS data. The continuous line represents a thermal component at the average cluster gas temperature of 7 keV (Markevitch et al. 1998). The errors bars are quoted at 1σ level.



All rights reserved Powered by 甜梦文库 9512.net

copyright ©right 2010-2021。