1: \documentclass[12pt,preprint]{aastex}
2:
3: \begin{document}
4:
5: \title{
6: S\'{e}rsic 159-03: discovery of the brightest soft X--ray excess emitting cluster of galaxies
7: }
8: \author{Massimiliano Bonamente$\,^{1}$,Richard~Lieu$\,^{1}$ and
9: Jonathan~P.~D.~Mittaz$\,^{2}$}
10:
11: \affil{\(^{\scriptstyle 1} \){Department of Physics, University of Alabama,
12: Huntsville, AL 35899, U.S.A.}\\
13: \(^{\scriptstyle 2} \){Mullard Space Science Laboratory, UCL,
14: Holmbury St. Mary, Dorking, Surrey, RH5 6NT, U.K.}\\
15: }
16:
17: \begin{abstract}
18: The soft X-ray excess emission in the southern cluster S\'{e}rsic 159-03
19: represents hiterto the strongest effect of its kind.
20: Emission in the $\sim 0.2-0.4$ keV passband is detected
21: far in excess of the expected contribution from
22: the hot phase of the intra-cluster medium, and
23: extends to the X-ray signal limit of the
24: cluster.
25: Our analysis of {\it ROSAT} PSPC observations reveal that the {\it soft excess} can be
26: interpreted as either a thermal or non-thermal effect, and
27: the high data quality allows to place tight constraints on the
28: two currently competing models.
29: However,
30: each model now
31: implies major revisions to our understanding of clusters of galaxies: either
32: `warm' gas masses in similar amounts to the hot gas, or relativistic particles
33: in or above equipartition with the hot phase, appear to be
34: unavoidable.
35: \end{abstract}
36:
37: \keywords{galaxies:clusters:individual (S\'{e}rsic 159-03) -- intergalactic medium}
38:
39: \section*{Introduction}
40: During the five intervening years since the discovery of EUV and
41: soft X--ray excess
42: emission in the Virgo and Coma clusters with the {\it Extreme Ultraviolet
43: Explorer} (EUVE) and {\it ROSAT} missions (Lieu et al. 1996a,b), the original results have been
44: cross-checked by follow-up observations of the same objects plus a larger sample of
45: nearby galaxy clusters (Fabian 1996; Bonamente, Lieu and Mittaz. 2001a,b; Bonamente et al. 2001c;
46: Mittaz, Lieu and Lockman 1998; Lieu
47: et al. 1999; Lieu, Bonamente and Mittaz 1999,2000; Bergh\"{o}fer, Bowyer and Korpela 2000a,b;
48: Bowyer and Bergh\"{o}fer 1998; Bowyer, Korpela and Bergh\"{o}fer 2001; Bowyer, Bergh\"{o}fer
49: and Korpela 1999; Arabadjis and Bregman 1999; Reynolds et al. 1999; Kaastra et al. 1999;
50: Dixon, Sallmen and Hurwitz 2001; Dixon, Hurwitz and Ferguson 1996;
51: Valinia et al. 2000; Fusco-Femiano et al. 2000
52: Buote 2000a,b,2001; Arnaud et al. 2001), using complementary data from the
53: {\it BeppoSAX}, {\it HUT}, {\it FUSE} and {\it RXTE} missions.
54: Recently, we have undertaken a more extensive study of soft X-ray emission in galaxy clusters
55: of low/intermediate $N_H$, with the aim of assessing the impact and cosmological significance
56: of the {\it soft excess} phenomenon. Our efforts led to the discovery of soft excess emission
57: in a sample of clusters in the Shapley concentration (Bonamente et al. 2001b), and while
58: the analysis of more sources is still underway, we report in this {\it Letter} the
59: brightest soft X-ray ($\sim 0.2-0.4$ keV) excess cluster within the {\it ROSAT} PSPC database.
60: The excess emission reaches $\sim$ 100 \% of the hot intra--cluster medium
61: (ICM) contribution, 3-4 times higher than the same for any other PSPC targets
62: (e.g., A1795 and Virgo; Bonamente, Lieu and Mittaz 2001a).
63: The inverse-Compton (IC) interpretation of the soft emission (see section 3) results
64: in a non-thermal pressure that greatly exceeds that of the thermal gas: the excess of S\'{e}rsic 159-03
65: has more demanding energetic requirements than all previously known cases (e.g.,
66: Coma, Virgo and A1795; Lieu et al. 1999, Bonamente, Lieu and Mittaz 2001a).
67: Alternatively, if the origin is thermal,
68: the mass implications of the putative `warm' gas are also
69: extreme, with the warm-to-hot gas mass ratio $\sim$ 50 \% (this figure is
70: matched only by the emission seen by EUVE for A2199, however at lower statistical significance;
71: Lieu, Bonamente and Mittaz 2000).
72:
73: \section*{X-ray, infrared and 21 centimeter observations of S\'{e}rsic 159-03}
74: S\'{e}rsic 159-03 is a rich southern galaxy cluster,
75: also known as Abell S1101,
76: at a redshift of z=0.056 (Abell, Corwin and Olowin 1989).
77: The cluster's X-ray luminosity (L$_X = 5.35 \times 10^{44}$ erg/s in 0.5-2.0 keV,
78: De Grandi et al. 1999) is typical for a cluster of its temperature (e.g., Wu, Xue and Fang 1999),
79: and the emission has a radial extent of $\sim$ 10 arcmin (Fig. 1). Recent XMM observations
80: (Kaastra et al. 2001) show a `cooling' of the hot ICM in the central regions and
81: no obvious peculiarities in the X-ray morphology, consistent with our PSPC data.
82: The cluster, moreover, has no known radio emissions associated with it.
83:
84: By means of three pointed PSPC observations of this source, taken between May 1992 and
85: May 1993 with a total
86: exposure of about 19,000 seconds, we found a very bright excess emission
87: in PSPC's 1/4 keV band (also known as R2 band, $\sim 0.2-0.4$ keV). Fig. 2 shows how
88: detected emission (light blue, dark blue and black crosses) compare with the
89: expected emission from the hot intra--cluster medium (ICM, solid lines), e.g. as described
90: by the recent X--ray analysis of Kaastra et al. (2001).
91: The three observations were modeled simultaneously, in order
92: to reduce statistical uncertainties.
93:
94: As extragalactic soft X--ray emissions are absorbed by the interstellar medium,
95: an accurate measurement of the line-of-sight
96: Galactic hydrogen (HI) column density ($N_H$) is crucial. The Dickey and Lockman (1990) HI maps
97: at resolutions of respectively 1, 4 and 9 square degrees centered at the cluster's position
98: indicate a region of sky free from spatial HI gradients, with
99: an HI density at the cluster position of 1.79 $\times 10^{20}$ cm$^{-2}$.
100: Since Galactic infrared 100 $\mu$m emission correlates well with HI 21-cm radiation
101: (Boulanger and Perault 1988),
102: we further consulted {\it IRAS} 100 $\mu$m maps (Wheelock et al. 1994)
103: at a finer resolution of 3 arcmin.
104: The 100 $\mu$m emission is very smooth over the cluster's region,
105: and it confirmed our $N_H$ value for the cluster's Galactic HI.
106: This value is henceforth used in the spectral analysis~\footnote{The low declination ($\sim -42^o$)
107: renders this source not easily observable with the NRAO 43-mt Green Bank telescope we employed
108: in our previous analyses.}. We refer to a previous paper (Bonamente et al. 2001b)
109: for the details on PSPC data analysis techniques.
110:
111: Spectra of azimuthally averaged annular regions were modelled with a single temperature
112: optically-thin plasma emission.~\footnote{The
113: spectral analysis was performed with the XSPEC software.
114: %Spectral models are MEKAL
115: %for the emissivity, and WABS for photoelectric absorption. The latter in fact employs
116: %the most reliable and up-to-date cross--sections, see Bonamente et al. 2001a
117: % for a discussion and references.
118: PSPC pulse-invariant (PI) channels
119: 1-19 and 202-256 are not used, as potentially affected by calibration uncertainties.}
120: The Galactic absorption was modelled with the codes of Morrison and MacCammon 1983 (MM83; `WABS' in XSPEC
121: language) and of Wilms, Allen and McCray 2000 (WAM00; `TBABS' in XSPEC). The He cross-sections
122: of MM83 are in good agreement with the recent compilation by Yan, Sadeghpour and Dalgarno (1998), as
123: also shown in a previous paper (Bonamente et al. 2001a). The WAM00 code provides
124: however the most up-to-date
125: cross-section compilation for absorption of X-rays in the interstellar medium.
126: In the energy
127: range of interest, 0.2-0.4 keV, the two codes are nearly indistinguishable at the
128: resolution of the PSPC (see Fig. 4 of
129: WAM00), and yielded statistically consistent results, see Tables 1 and 2. The emission
130: code employed was MEKAL (after Mewe, Gronenschild and van den Oord, 1985;
131: Mewe, Lemen and van den Oord 1986; Kaastra 1992).
132:
133: When applied to the entire passband of the PSPC (0.2-2.0 keV),
134: the fit is formally unacceptable
135: (e.g., red crosses in Fig. 3), and a region of depleted
136: flux (0.5-1.0 keV) appears to accompany the 0.2-0.4 keV excess. The fit is satisfactory
137: when applyed only to energies $\geq$ 0.5 keV, with the soft excess
138: then abundantly revealed at E$\leq$ 0.4 keV (green crosses, Fig. 3). The best-fit parameters
139: of the hot phase agree with the most recent X--ray
140: {\it XMM} observation (Kaastra et al. 2001).
141: In Table 1 we
142: summarise the results.
143: We emphasize that the soft excess emission is so strong that, in order to be interpreted as
144: the low--energy tail of the hot ICM emission, a Galactic column density as low as
145: 5-8 $\times 10^{19}$ cm$^{-2}$ would be required, which is not only severely discrepant
146: from the
147: measured
148: values, but even lower that that of the `Lockman hole' where $N_H$ reaches global minimum.
149: The soft emission cannot therefore be attributed to peculiarities in the Galactic
150: HI distribution.
151:
152: In Figure 4 we show the trend of the
153: PSPC excess. The
154: emission extends to a radius of 800 kiloparsec~\footnote{A Hubble constant
155: of $H_0=50$ km s$^{-1}$ Mpc$^{-1}$ is assumed here and after.} (=9 arcmin); in the 9--12 arcmin
156: region, where only marginal cluster emission is detected (Fig. 1), the data point
157: is consistent with zero flux, thereby confirming the integrity of our
158: background subtraction technique (Bonamente et al. 2001b for reference).
159: As comparison, we note that from the {\it XMM}/EPIC data
160: of Kaastra et al. (2001) evidence for X-ray emission
161: is apparent out to a radial distance of about 10 arcmin, in agreement
162: with our PSPC data. The soft excess
163: emission therefore persists to the X-ray detection limit.
164: To illustrate the significance of the role played by the model
165: uncertanties in our determination of the 1/4 keV contribution from the hot ICM,
166: we consider as example the 3-6' region. When the statistical error in the
167: 0.5-2.0 keV PSPC data (the portion of the spectra used
168: for model determination) is taken into account, the error in the fractional eccess
169: increases only
170: from 9 \% (see Fig. 4) to 10\%.
171: Similar conclusions apply to the other regions, indicating that
172: evidence for a soft X-ray component from
173: S\'{e}rsic 159-03 is unaffected by statistical uncertanties in the hot ICM modelling
174: ~\footnote{Krick, Arabadjis and Bregman (2000)
175: argue that sub-arcmin variations of $N_H$ could in
176: principle account for a 20-30 \% excess. Although the relevance of this phenomenon
177: to our S\'{e}rsic 159-03 observations is uncertain, the strength of the PSPC excess here reported
178: largely exceeds that figure. On occasions where comparison with stellar Lyman-$\alpha$
179: and quasar X-ray spectra could be made, the agreement among the different methods of
180: determining $N_H$ would suggest an error less than $\sim 10^{19}$ cm$^{-2}$
181: (see Lieu et al. 1996b and references therein).}.
182:
183: \section*{Interpretation}
184:
185: How do we interpret such a strong signal?
186: A `warm', optically--thin phase of the intra-cluster medium (ICM)
187: at sub-megakelvin temperatures can in principle accont for the
188: excess emission.
189: Thermal conduction can be suppressed, either for the presence of tangled magnetic fields
190: (Chandran et al. 1999) or if the warm phase is sufficiently `clumpy'
191: (Cowie and McKee 1977; Smith and Lilliequist 1979),
192: yet the time scale for radiative losses of a
193: `warm' gas
194: is sufficiently short to necessitate rapid and continuous replenishment
195: (such as that of the `mixing layers' scenario of Fabian 1997), if the
196: warm phase existed in a steady state for cosmological times.
197: In the neighborhood of T $\sim 10^6$ K, the cooling time of a gas with solar abundances
198: is $t_{cool} \sim 3 \times 10^7 \times n_{-3}^{-1}$ years, where $n_{-3}$
199: is gas density in units of $10^{-3}$ cm$^{-3}$ (Landini and Monsignori-Fossi 1990),
200: the characteristic density of the warm phase in the 1-3' and 3-6' regions for a
201: volume filling factor $f=1$ ( 0 $<$ $f \leq$ 1).
202: Our observations however determine the so called {\it emission measure} of the warm phase
203: ($EM \propto \int n^2 dV$, $V$ being the volume of the emitting gas),
204: and $f$ remains unknown.
205: For a given detected $EM$, the total warm gas mass $M$ in a given region is
206: a function of $f$, $M \propto \sqrt{f}$;
207: as the density increases with decreasing $f$ ($n \propto 1/ \sqrt{f}$),
208: the cooling time accordingly decreases as $t_{cool} \propto \sqrt{f}$.
209: The mass implications are very demanding: this new component
210: will have a mass comparable to that of the hot ICM, see Table 2, and those
211: budgets can
212: be alleviated if $f$ is very small, however at the price of a correspondingly shorter
213: cooling time.
214:
215: Alternatively, an inverse Compton (IC) origin of the emission can be advocated (Sarazin and Lieu 1998).
216: Diffusive shock acceleration (Axford et al. 1977; Bell 1978a,b; Blandford
217: and Ostriker 1978) may be responsible for a population of
218: relativistic electrons with a power-law spectrum in the ICM, and
219: its IC mechanism preserves a similar power-law shape, whereby the two differential
220: number distributions of electrons (N(E)$\propto E^{-\mu}$) and emitted
221: photons (L$(\epsilon)
222: \propto \epsilon^{-\alpha}$) are related by the equation $\alpha=(1+\mu)/2$.
223: Relativistic electrons of $\gamma$ =300-700,
224: required to emit EUV or soft X--ray through IC scattering,
225: straddle spectrally between low--energy Coulomb losses and high--energy radiative losses
226: (Lieu et al. 1999; Sarazin 1999),
227: and may survive for a significant fraction of a cluster's lifetime.
228: Here we fit the photon index $\alpha$
229: to the data, Table 2, with
230: the exception of the 3-6 arcmin region
231: where $\alpha$ it is fixed at $\alpha$=1.75, the expected value from
232: a cosmic ray (CR) population of $\mu$=2.5 in accordance to
233: Galactic CR.
234: The modelling is formally acceptable, although with a spectral index somewhat
235: steeper than that of Galactic CR.
236: However, serious problems confront the energetic requirements of the IC model: the pressure
237: of CR electrons with $\gamma$ in the range 475--700 (those which emit IC radiation
238: in the E=0.2-0.4 keV band) exceeds that of the hot ICM by a factor of order 2-3!
239: Moreover, this pressure estimate is necessarily a lower limit, as
240: higher-- and lower--energy electrons may be present, as well as CR ions which, although not
241: visible, will further drive the CR budget to absurde values (Lieu et al. 1999; see also Miniati
242: et al. 2001).
243:
244: As often happens in rich clusters of galaxies, the X--ray centroid, here coincident with the
245: central cD galaxy ESO 291-9, is associated with a lower temperature for
246: the hot ICM (Kaastra et al. 2001).
247: Such a temperature decrease is also clearly visible in our PSPC data.
248: Yet the detected soft X--ray excess cannot be directly related to it: firstly, because
249: the excess covers an area about 25 times larger than the region affected by this `cooling'
250: (which has a radius of about 1.8 arcmin, Kaastra et al. 2001; Allen and Fabian 1997);
251: secondly, and more importantly,
252: because we already accounted for the effect in our data modeling (see best--fit
253: temperatures in Table 1).
254: The behaviour here is therefore analogous to, e.g., that of the Virgo and A1795 clusters,
255: where the excess emission spreads over a much larger area than that of
256: the central cooler region (Lieu et al. 1996a; Bonamente et al. 2001a).
257: The central galaxy of S\'{e}rsic 159-03, which also contains an IR source,
258: is well contained within a radius of 1 arcmin (Hansen et al. 2000): the soft excess emission
259: must then be a genuinely cluster--wide phenomenon.
260:
261: \section*{Discussion and conclusions}
262: The high statistical significance of the soft excess emission of this cluster
263: confounds both thermal and non-thermal models,
264: Table 2. Either model can fit the data,
265: and calls for an overhaul in
266: our understanding of galaxy clusters.
267:
268: Following the non-thermal interpretation, the large pressure budgets of
269: Table 2
270: cannot easily be mitigated; on the contrary, they are strict lower limits
271: to the relativistic particle content in the cluster.
272: The presence of a population of relativistic particles
273: {\it above} equipartition with the hot gas
274: clearly has implications on a cluster's hydrostatic balance
275: (e.g., Berezinsky et al. 1997), and the role of intra-cluster
276: CR on the cluster's evolution (e.g., heating of the ICM gas)
277: could be much more significant than previously thought. At present it is not known
278: whether strong shocks, e.g., those induced by cluster mergers, may accelerate
279: such a large amount of CRs. It is possible that the cluster experiences many such
280: shocks throughout its lifetime; the ICM would then accumulate
281: those
282: CR electrons with the longest radiative life time, precisely
283: the $\gamma \sim$ 300-500 electrons which emit in EUV and soft X-rays through
284: IC scattering (Sarazin
285: and Lieu 1998; Lieu et al. 1999).
286:
287: On the other hand, large warm gas masses could be sustained
288: at the interface of cold gas clouds and the hot ICM (Fabian 1997),
289: the presence of a `cold' (T $\leq 10^4$ K) phase in the ICM
290: is in fact an interpretation of the PSPC images of the Coma and Virgo clusters
291: (Bonamente et al. 2001c). In the absence of such a replenishment mechanism, however,
292: the very short radiative cooling time renders the thermal model untenable.
293:
294: The bright soft excess of S\'{e}rsic 159-03 here reported
295: presents a new and unavoidable reality: whether the ultimate explanation be thermal,
296: non-thermal or some other origin, a
297: major effect at work in the intergalactic medium has hiterto been
298: completely ignored.
299:
300: \newpage
301:
302: \section*{References}
303:
304: \noindent
305: Abell, G.O., Corwin, H.G. and Olowin, R.P. 1989, {\it ApJS}, {\bf 70}, 1.\\
306: Allen, S.W. and Fabian, A.C. 1997, {\it MNRAS}, {\bf 286}, 583. \\
307: \noindent
308: Anders, E. and Grevesse, N. 1989, {\it Geochim. Cosmochim. Acta}, {\bf 53}, 197. \\
309: \noindent Arabadjis J.S. \& Bregman, J.N., 1999, {\it ApJ},
310: {\bf 514}, 607. \\
311: \noindent
312: Arnaud, M. et al. 2001, {\it A \& A}, {\bf 365}, L80.\\
313: \noindent
314: Axford, W.I., Leer, E. and Skadron, G. 1977, {\it Proc. 15th Int. Cosmic-Ray \\
315: \indent Conference (Plovdiv)}, {\bf 11}, 132. \\
316: \noindent
317: Bell, A.R. 1978a, {\it MNRAS}, {\bf 182}, 147.\\
318: \noindent
319: Bell, A.R. 1978b, {\it MNRAS}, {\bf 182}, 443.\\
320: \noindent
321: Berezinski, V.S., Blasi, P. and Ptuskin, V.V 1997, {\it ApJ}, {\bf 487}, 529.\\
322: \noindent
323: Berghoefer, T.W., Bowyer, S., \& Korpela, E.J., 2000, {\it ApJ},
324: {\bf 545}, 695. \\
325: \noindent
326: Berghoefer, T.W., Bowyer, S., \& Korpela, E.J., 2000, {\it ApJ},
327: {\bf 535}, 615. \\
328: \noindent
329: Blandford, R.D. and Ostriker, J.P. 1978, \it A \& A \rm, {\bf 221}, L29. \\
330: \noindent
331: Bonamente, M., Lieu, R. and Mittaz, J.P.D. 2001a, \it ApJ \rm, {\bf 547}, L7. \\
332: \noindent
333: Bonamente, M., Lieu, R. , Nevalainen, J. and Kaastra, J.S. 2001b, \\
334: \indent \it ApJ\rm, {\bf 552}, L7. \\
335: \noindent
336: Bonamente, M., Lieu, R. and Mittaz, J.P.D. 2001c, \it ApJ \rm, {\bf 546}, 805.\\
337: \noindent
338: Boulanger, F. and Perault, M. 1988, {\it ApJ}, {\bf 330}, 964. \\
339: \noindent
340: Bowyer, S., Berghoeffer, T.W. 1998, {\it ApJ}, {\bf 506}, 502. \\
341: \noindent
342: Bowyer, S., Berghoeffer, T.W., \& Korpela, E.J., 1999, {\bf ApJ},
343: {\it 526}, 592. \\
344: \noindent
345: Bowyer, S., Korpela, E.J., \& Berghoefer, T.W., 2001, {\it ApJ}, {\bf 548},
346: L135. \\
347: \noindent
348: Buote, D.A., 2000, {\it ApJ}, {\bf 544}, 242.\\
349: \noindent
350: Buote, D.A., 2000, {\it ApJ}, {\bf 532}, L113.\\
351: \noindent
352: ~Buote, D.A., 2001, {\it ApJ}, {\bf 548}, 652. \\
353: \noindent
354: Chandran, B.D.G., Cowley, S.C and Allbright, B. 1999, {\it Diffuse thermal and \\
355: \indent relativistic plasma in galaxy clusters}, MPI, Garching, 242. \\
356: \noindent
357: Cowie, L.L. and McKee, C.F. 1977, {\it ApJ}, {\bf 211}, 135. \\
358: \noindent
359: De Grandi, S. et al. 1999, {\it ApJ}, {\bf 514}, 148. \\
360: \noindent
361: Dickey., J.M. and Lockman, F.J. 1990, {\it Ann. R. Astron. Astrop.},
362: {\bf 28}, 215. \\
363: \noindent
364: Dixon, W.V., Sallmen, S., Hurwitz, M., \& Lieu, R., 2001,
365: {\it ApJ}, {\bf 550}, L25.\\
366: \noindent
367: Dixon, W.V., Hurwitz, M., \& Ferguson, H.C., 1996, {\it ApJ},
368: {\bf 469}, L77. \\
369: \noindent Fabian, A.C. 1996, {\it Science}, {\bf 271}, 1244. \\
370: \noindent Fabian, A.C. 1997, {\it Science}, {\bf 275}, 48. \\
371: \noindent
372: Fusco-Femiano, R., Dal Fiume, D., et al, 2000, {\it ApJ}, {\bf 534}, L7. \\
373: \noindent
374: Hansen, H.E., J\o rgensen. H.E., N\o rgaard-Nielsen, H.U, Pedersen, K., \\
375: \indent Goudfrooij, P and Linden-V\o rnle, M.J.D. 2000, {\it A \& A}, {\bf 362}, 1018.\\
376: \noindent
377: ~Kaastra, J.S. 1992 in \it An X-Ray Spectral Code for Optically Thin Plasmas \rm
378: \\\indent
379: (Internal SRON-Leiden Report, updated version 2.0). \\
380: \noindent
381: Kaastra, J.S., Lieu, R., Mittaz, J.P.D. et al. 1999, {\it ApJ}, {\bf 519}, L119. \\
382: \noindent
383: Kaastra, J.S., Ferrigno, C., Tanura, T., Paerels, F.B.S., Peterson, J.R. \\
384: \indent and Mittaz, J.P.D. 2001, \it A \& A \rm, {\bf 365}, 99. \\
385: \noindent
386: Landini, M. and Monsignori-Fossi B.C. 1990, {\it A \& AS}, {\bf 82}, 229.\\
387: \noindent
388: Krick, J., Arabadjis, J.N. and Bregman, J.N. 2000, {\it BAAS}, {\bf 197}, 07.11. \\
389: \noindent
390: Lieu, R., Mittaz, J.P.D., Bowyer, S., Lockman, F.J.,
391: Hwang, C. -Y., Schmitt, \\
392: \indent J.H.M.M. 1996a, \it ApJ \rm, {\bf 458}, L5. \\
393: \noindent
394: Lieu, R., Mittaz, J.P.D., Bowyer, S., Breen, J.O.,
395: Lockman, F.J., \\
396: \indent Murphy, E.M. \& Hwang, C. -Y. 1996b, {\it Science}, {\bf 274},1335--1338. \\
397: \noindent
398: Lieu, R., Bonamente, M., Mittaz, J.P.D., Durret, F., Dos Santos, S. and \\
399: \indent Kaastra, J. 1999a, ApJ, {\bf 527}, L77.\\
400: \noindent
401: Lieu, R., Bonamente, M. and Mittaz, J.P.D. 1999,
402: {\it ApJ}, {\bf 517}, L91.\\
403: \noindent Lieu, R., Bonamente, M., \& Mittaz, J.P.D., 2000, {\it A \& A}, {\bf 364},
404: 497.\\
405: \noindent
406: Lieu, R., Ip, W.-I., Axford, W.I. and Bonamente, M. 1999, {\it ApJ}, {\bf 510},
407: L25.\\
408: \noindent
409: ~Mewe, R., Gronenschild, E.H.B.M., and van den Oord, G.H.J., 1985 \\
410: \indent {\it A\&A Supp.}, {\bf 62}, 197. \\
411: \noindent Mewe, R., Lemen, J.R., and van den Oord, G.H.J. 1986,
412: \it A\&A. Supp.\rm, \\
413: \indent {\bf 65}, 511--536. \\
414: \noindent
415: Miniati, F., Ryu, D., Kang, H. and Jones, T.W. 2001, \it ApJ \rm in press. \\
416: \noindent
417: Mittaz, J.P.D., Lieu, R. and Lockman, F.J. 1998, \it ApJ\rm, 498, L17. \\
418: \noindent
419: Reynolds, A.P et al. 1999,
420: {\it A \& AS}, {\bf 134}, 287.\\
421: \noindent
422: Morrison, R. and McCammon, D. 1983, {\it ApJ}, {\bf 270}, 119.\\
423: \noindent
424: Sarazin, C.L. and Lieu, R. 1998, {\it ApJ}, {\bf 494}, L177. \\
425: \noindent
426: Sarazin, C.L. 1999, {\it ApJ}, {\bf 520}, 529.\\
427: \noindent
428: Smith, D.F. and Lilliequist, C.G. 1979, {\it ApJ}, {\bf 232}, 582. \\
429: \noindent Valinia, A. Arnaud, K., Loewenstein, M., et al. 2000,
430: {\it ApJ}, {\bf 541}, 550.\\
431: \noindent Yan, M., Sadeghpour, H.R. and Dalgarno, A. 1998, {\it ApJ}, {\bf 496}, 1044.\\
432: \noindent
433: Wilms, J., Allen, A. and MCCray R. 2000, {\it ApJ}, {\bf 542}, 914.\\
434: \noindent
435: Wheelock, S. et al., 1994, {\it IRAS Sky Survey Expl. Supp.}, (IPL Publ. 94-11), \\
436: \indent Pasadena, CA. \\
437: \noindent
438: Wu, X.-P., Xue, Y.-J. and Fang, L.-Z. 1999, {\it ApJ}, {\bf 524}, 22. \\
439:
440: \newpage
441:
442: \begin{table}[h!]
443: \begin{center}
444: \tighten
445: \caption{Spectral modelling of PSPC spectra of S\'{e}rsic 159-03 with the simple
446: 1-T model. Here and after errors are 90 \% confidence ($\chi^2$ + 2.7 method),
447: where no errors are reported parameters were held fixed, according to the X--ray study of
448: Kaastra et al. (2001). Fit is performed by minimizing the $\chi^2$ function, d.o.f. is degrees
449: of freedom. `A' is elemental abundance following the relative proportions of Anders and Grevesse
450: (1989). Photoelectric absorption code is MM83, in square brackets the new WAM00
451: code is employed, with same results as MM83.
452: }
453: \small
454: \vspace{2cm}
455: \hspace{-1cm}
456: \begin{tiny}
457: \begin{tabular}{lcccccccccc}
458: \hline
459: Region & \multicolumn{3}{c}{0.2-2.0 keV fit} &\multicolumn{4}{c}{0.2-2.0 keV fit, free NH} &
460: \multicolumn{3}{c}{0.5-2.0 keV fit} \\
461: (arcm.) & k$T$ & A & $\chi^2$(d.o.f) & k$T$ & A & NH & $\chi^2$(d.o.f) & k$T$ & A & $\chi^2$(d.o.f) \\
462: & (keV) & & (keV)& & $10^{20}$ cm$^{-2}$ & & & (keV) & & \\
463: & \multicolumn{3}{c}{\hrulefill} & \multicolumn{4}{c}{\hrulefill} & \multicolumn{3}{c}{\hrulefill} \\
464: 0-1 & 1.44 $\pm^{0.1}_{0.02}$ & 0.08 $\pm^{0.02}_{0.02}$ & 264(167) &
465: 1.88 $\pm^{0.23}_{0.11}$ & 0.4 $\pm^{0.12}_{0.09}$ & 0.95 $\pm^{0.12}_{0.11}$ & 143(166)&
466: 1.87 $\pm^{0.14}_{0.11}$ & 0.41 $\pm^{0.14}_{0.1}$ & 97(137) \\
467: & [ 1.48 $\pm 0.07$ & 0.08 $\pm^{0.015}_{0.02}$& 272(167) &
468: 1.92 $\pm^{0.12}_{0.14}$& 0.43 $\pm^{0.14}_{0.1}$ & 0.92 $\pm^{0.12}_{0.13}$ &142(166) &
469: 1.87 $\pm^{0.14}_{0.11}$ & 0.41 $\pm^{0.14}_{0.1}$ & 97(137) ] \\
470: 1-3 & 1.38 $\pm{0.08}$ & 0.03 $\pm^{0.02}_{0.03}$ & 335(162) &
471: 2.9$\pm^{0.9}_{0.4}$ & 0.5$\pm^{0.34}_{0.21}$ & 0.52 $\pm^{0.13}_{0.14}$ & 154(161) &
472: 2.9$\pm^{0.7}_{0.5}$ & 0.5$\pm^{0.37}_{0.23}$ & 115(132) \\
473: & [ 1.38 $\pm^{0.08}_{0.1}$ & 0 $\pm^{0.05}_{0}$ & 342(162) &
474: 3 $\pm^{0.8}_{0.5}$ & 0.5 $\pm^{0.36}_{0.21}$ & 0.51 $\pm^{0.14}_{0.15}$ & 153(161) &
475: 2.9 $\pm{0.65}$ & 0.5 $\pm^{0.38}_{0.23}$ & 115(132) ]\\
476: 3-6 & 2.0$\pm^{0.5}_{0.36}$ & 0.02 $\pm^{0.08}_{0.02}$ & 111(128) &
477: 4.0$\pm^{3.8}_{1.4}$ & 0.6$\pm^{6}_{0.5}$ & 0.7$\pm^{0.3}_{0.5}$ & 88(127) &
478: 3.8$\pm^{4}_{1.4}$ & 0.6$\pm^{5}_{0.5}$ & 66(99) \\
479: & [ 2 $\pm^{0.45}_{0.76}$ & 0 $\pm^{0.01}_{0}$ & 112(128)&
480: 4.1 $\pm^{3.8}_{1.4}$ & 0.77 $\pm^{5}_{0.6}$ & 0.68 $\pm^{0.32}_{0.45}$& 88(127) &
481: 3.7 $\pm^{1.9}_{1.2}$ & 0.64 $\pm^{5}_{0.63}$ & 66(99) ] \\
482: 6-9 & 2 & 0.3 & 67(84) & 2 & 0.3 & 1.3$\pm^{0.8}_{0.65}$ & 65(83) & 2 & 0.3 & 31(55) \\
483: &[ 2 & 0.3& 67(84) & 2 & 0.3 & 1.3 $\pm^{0.85}_{0.4}$ & 66(83) & 2 & 0.3 & 31(55) ] \\
484: \hline
485: \end{tabular}
486: \end{tiny}
487: \end{center}
488: \end{table}
489:
490: \newpage
491:
492: \begin{table}[h!]
493: \begin{center}
494: \tighten
495: \caption{Spectral modelling with a 2-T model (`hot' and `warm' phases of ICM) and with a
496: non--thermal model. Mass implications for the former and pressure budgets for the latter
497: are reported; $f$ is filling factor of the warm component. See Table 1 for explanation of
498: Galactic absorption codes employed.}
499: \vspace{2cm}
500: \hspace{-1cm}
501: \begin{tiny}
502: \begin{tabular}{lccccccccccc}
503: \hline
504: Region & \multicolumn{5}{c}{2-T model} & \multicolumn{6}{c}{Non-thermal model} \\
505: (arcm.) & k$T_h$ & A & k$T_w$ & $\chi^2$(d.o.f) & $M_w/M_h$ & k$T$ & A & $\alpha $ & $\chi^2$(d.o.f) &
506: $P_{therm}$ & $P_{IC}$ \\
507: & (keV) & & (keV) & & &(keV) & & & & \multicolumn{2}{c}{$10^{-11}$ (erg cm$^{-3})$} \\
508: & \multicolumn{5}{c}{\hrulefill} & \multicolumn{6}{c}{\hrulefill} \\
509: 0-1 & 1.88$\pm^{0.33}_{0.11}$ & 0.43$\pm^{0.32}_{0.07}$ & 0.078$\pm^{0.035}_{0.015}$ & 130(165) &
510: $\sqrt{f} \times$ 0.38$\pm^{0.25}_{0.05}$ & 1.94$\pm^{0.16}_{0.13}$ & 0.5 & 3.2$\pm^{0.9}_{0.2}$ &
511: 136(166) & 7 & 22$\pm^{25}_{13}$ \\
512: & [ 1.9 $\pm^{0.2}_{0.12}$ & 0.64$\pm^{00.15}_{0.11}$ & 0.07$\pm{0.035}$& 129(165)&
513: $\sqrt{f} \times$ 0.34 $\pm^{0.26}_{0.17}$ & 1.94$\pm^{0.16}_{0.13}$ & 0.5 & 3.2$\pm^{1.1}_{0.3}$ &
514: 135(166) & & 20 $\pm^{22}_{12}$ ] \\
515: 1-3 & 2.9 $\pm^{0.74}_{0.46}$ & 0.53$\pm^{0.47}_{0.21}$ & 0.06$\pm^{0.04}_{0.021}$& 142(160) &
516: $\sqrt{f} \times$ 0.55$\pm^{0.6}_{0.2}$ & 2.8$\pm^{0.8}_{0.4}$ & 0.4 & 4.8$\pm^{1.2}_{0.6}$ &
517: 148(161) & 0.7 & 2.3 $ \pm^{7.5}_{1.5} $ \\
518: &[ 3$\pm^{0.85}_{0.5}$ & 0.67 $\pm^{0.68}_{0.32}$ & 0.07 $\pm^{0.035}_{0.02}$ & 141(160) &
519: $\sqrt{f} \times$ 0.33 $\pm^{0.82}_{0.22}$ & 2.8$\pm^{0.8}_{0.4}$ & 0.4 & 4.7 $\pm^{1.2}_{0.2}$ &
520: 148(161) & & 2 $\pm^{7}_{1.3}$ ] \\
521: 3-6 & 3.5$\pm^{2.7}_{1.2}$ & 0.3 & 0.06$\pm{0.0035}$ & 88(127) &
522: $\sqrt{f} \times$ 0.46$\pm^{15}_{0.26}$ & 2.5 & 0.2 & 1.75 &
523: 96(129) & 0.14 & 0.33 $\pm^{0.1}_{0.11}$ \\
524: & [ 3.5 $\pm^{3.1}_{1.2}$ & 0.3 & 0.055$\pm^{0.06}_{0.03}$ & 88(127) &
525: $\sqrt{f} \times$ 0.61 $\pm^{17}_{0.42}$ & 2.5 & 0.2 & 1.75 &
526: 96(129) & & 0.36 $\pm{0.12}$ ] \\
527: \hline
528: \end{tabular}
529: \end{tiny}
530: \end{center}
531: \end{table}
532:
533: \newpage
534:
535: Figure 1: Radial profiles of X-ray brightness in PSPC R2 (0.2-0.4 keV, red) and R37
536: (0.4-2.0 keV, green)
537: bands in units of counts/s/sq.arcmin, the lines mark the corresponding background levels.
538: Background is determined from an annulus $\sim$ 40' off-axis.
539:
540: Figure 2: Three PSPC spectra of the central region (radius 3 arcmin) of S\'{e}rsic 159-03. The data,
541: available through the public archive (identification numbers 800397, 600182 and 600429a1), are
542: individually rebinned to obtain a minimum S/N $>$ 4 in each spectral channel. The solid lines describe the
543: best--fit single temperature model (see text for description) that highlights the low--energy
544: excess of photons. Agreement among the 3 observations is excellent, see bottom panel.
545:
546: Figure 3: Coadded spectrum of 1-3 arcmin region of the cluster (black crosses) overlaid
547: on the best fit 1-T model obtained by fitting the whole PSPC band (left, red solid line)
548: and by fitting only energies $\geq$ 0.5 keV (right, green solid line). Residuals (red and green crosses)
549: reveal in both cases excess of soft photons at energies $<$ 0.4 keV.
550:
551: Figure 4: Fractional excess in 1/4 keV PSPC band (0.2-0.4 keV) as function of radial
552: distance from the cluster's center. Fractional excess is defined as $\eta=(m-p)/p$, where $m$ is measured
553: soft flux (1/4 keV band) and $p$ its prediction according to the 0.5-2.0 keV single temperature model.
554: The 9-12' region was modelled with the same parameters of the 6-9' annulus (Table 1);
555: vertical semi-diameters are 1-$\sigma$ statistical errors.
556: The excess component ranges between 50 and 100 \% of the hot ICM gas in the soft band, the
557: strongest observed to date in PSPC data.
558:
559: %Figure 4: Contour plot of X-ray emission from S\'{e}rsic 159-03 overlaid on an
560: %optical image ({\it Digitized
561: %Sky Survey}) of the cluster region. Contours are 200,70,20,6.5,3 and 1.5 counts/pixel,
562: %estimated background is 0.45 counts/pixel.
563: %The central source in the optical is the cD galaxy ESO 291-9, also at the
564: %center of X--ray emission, other two bright features are foreground
565: %stars unrelated to the cluster.
566:
567:
568: \end{document}
569: