1: % iaus2esa.tex -- sample pages for Proceedings IAU Symposium document class

2: % (based on v1.0 cca2esam.tex)

3: % v1.04 released 17 May 2004 by TechBooks

4: % small changes and additions made by KAvdH/IAU 4 June 2004

5: % Copyright (2004) International Astronomical Union

6: 

7: \NeedsTeXFormat{LaTeX2e}

8: 

9: \documentclass{iaus238}

10: \usepackage{graphicx}

11: \newcommand{\msun}{M_{\odot}}

12: 

13: \title[Soft X-ray spectrum of PG1211+143] 

14: {The soft X-ray spectrum of PG1211+143}

15: 

16: \author[K.~A. Pounds]   

17: {K.~A. Pounds,$^1$ K.~L. Page$^1$ \and J.~N. Reeves$^2$}

18: 

19: \affiliation{$^1$~Department of Physics and Astronomy, University of Leicester, UK\\[\affilskip]

20: $^2$~NASA Goddard Space Flight Center, Code 663, Greenbelt, MD 20771, USA \\ }

21: 

22: \pubyear{2006}

23: \volume{238}  

24: \pagerange{001--999}

25: \date{??? and in revised form ???}

26: \setcounter{page}{001}

27: \jname{Black Holes: from Stars to Galaxies -- across the Range of Masses}

28: \editors{V. Karas \& G. Matt, eds.}

29: \begin{document}

30: 

31: \maketitle

32: 

33: \begin{abstract}

34: The narrow line QSO PG1211+143 has been a focus of recent attempts to

35: understand the soft excess in AGN, while the 2001 {\it XMM-Newton}

36: observation of this luminous AGN also provided evidence for a massive

37: and energetic outflow. Here we consider a physical link between the

38: energetic outflow and the variable soft excess.

39: \keywords{Galaxies: active -- galaxies: nuclei}

40: \end{abstract}

41: 

42: \firstsection % if your document starts with a section,

43:               % remove some space above using this command.

44: \section{Introduction}

45: 

46: One of the most important new results emerging from {\it XMM-Newton} and

47: {\it Chandra} observations of AGN is the evidence for sub-relativistic

48: outflows from a number of  luminous quasars.  For two of these,

49: APM08279+5255 (Chartas \etal\ 2002) and PG1115+080 (Chartas \etal\

50: 2003), both BAL quasars, the immediate inference was to associate the

51: X-ray columns with the high velocity gas seen in optical and UV spectra

52: of these unusual objects. In such cases the high columns might result

53: from viewing the continuum source through a wind being blown

54: tangentially off the accretion disc (Proga \etal\ 2000).

55: 

56: However, in 2 further cases, PG1211+143 (Pounds \etal\ 2003) and PDS456

57: (Reeves \etal\ 2003), a moderate and a high luminosity QSO,

58: respectively, evidence of a high velocity -- and hence energetic --

59: outflow suggested this could be a common property of luminous, or

60: perhaps more specifically, of high accretion rate quasars (King and

61: Pounds 2003). Unless highly collimated, such outflows will form a

62: substantial  component of the mass and energy budgets in luminous AGN,

63: while having important implications for metal enrichment of the

64: intergalactic medium and for the feedback mechanism implied by the

65: correlation of black hole and galactic bulge masses (King 2005).

66: 

67: The {\it XMM-Newton} observation of PG1211+143 in 2001 has also been

68: used recently to explore the physical origin of the `soft excess' in the

69: X-ray spectra of many AGN, following doubts raised about Comptonisation

70: models (Gierlinski and Done 2004). Alternative descriptions have been

71: proposed, invoking strong reflection  from a highly ionised accretion

72: disc (Crummy \etal\ 2006) and absorption in high velocity ionised gas

73: (Schurch and Done 2006, Chevallier \etal\ 2006). Here we also examine

74: EPIC data from a second observation of PG1211+143 in 2004, where a

75: conventional plot shows a weaker but `hotter' soft excess (Figure 1L).

76: 

77: A direct comparison of the EPIC data from 2001 and 2004 (Figure 1R)

78: suggests the spectral difference is primarily due to a variable

79: `deficit' of flux at $\sim$0.7--2 keV, supporting the contention that the

80: spectral curvature near $\sim$1 keV is indeed an artefact of absorption

81: on a  steeper underlying continuum. However, if the flux deficit is due

82: to photoionised absorption,  it is interesting that the hard X-ray flux

83: -- which should be a good proxy for the ionising luminosity -- is

84: essentially unchanged, suggesting that a reduced covering factor (or

85: column density) is the cause of the weaker `soft excess' in 2004.  Here

86: we suggest an alternative possibility (explored more fully in Pounds and

87: Reeves 2006), that a variable continuum component partly `fills in' the

88: absorption flux deficit in the 2004 spectrum. While contributing

89: directly to the strong soft X-ray flux in PG1211+143, the  additional

90: soft continuum component also explains the lack of strong `narrow'

91: absorption lines in the soft X-ray band (Kaspi and Behar 2006), in

92: conflict with the relatively strong, blue-shifted absorption lines  in

93: the EPIC data (Pounds and Page 2006). 

94: 

95: We speculate that the additional, steep power law continuum is powered by

96: the energetic high velocity outflow. 

97: 

98: \begin{figure}

99: \rotatebox{-90}

100: {\includegraphics[width=.18\textheight]{praguefig1.ps}}

101: \rotatebox{-90}

102: {\includegraphics[width=.18\textheight]{praguefig2.ps}} 

103: \caption{Left panel: conventional plot of the 2001 (black) and 2004

104: (grey) EPIC pn spectra showing a strong soft excess sitting above the

105: hard power law. On this view it appears that the soft excess in 2004 is

106: weaker but `hotter'.  Right panel: direct comparison of the

107: background-subtracted EPIC data  showing the spectral difference  to be

108: due to an increase in flux at $\sim$0.7--2 keV and a (less obvious, but

109: significant) decrease at $\sim$0.4--0.7 keV in 2004 (grey).}

110: \label{l}

111: \end{figure}

112: 

113: \section{Confirming an energetic outflow in PG1211+143}

114: PG1211+143 (at z=0.0809) is a prototypical `Big Blue Bump' quasar, and

115: one of the brightest AGN in the soft X-ray band (Elvis \etal\ 1991). It

116: is also classified (Kaspi \etal\ 2000) as a Narrow Line Seyfert 1 galaxy

117: (FWHM H$\beta$ 1800 km~s$^{-1}$), with a small central black hole mass 

118: ($\sim4\times10^{7}$M$_{\odot}$) for an object of bolometric luminosity

119: $4\times10^{45}$erg~s$^{-1}$, implying that the accretion rate is close

120: to  the Eddington  limit.  

121: 

122: The initial $\sim$50 ks observation of PG1211+143 with XMM-Newton in

123: 2001 revealed several blue-shifted absorption lines, interpreted by

124: Pounds \etal\ (2003) as evidence for a highly ionised outflow with a

125: velocity v$\sim$0.09c, a conclusion primarily based on identifying the

126: strongest $\sim$7 keV feature with FeXXVI Ly$\alpha$. As Kaspi and Behar

127: (2006) were unable to confirm this result in a careful ion-by-ion

128: modelling of the RGS data we have re-examined the higher signal-to-noise

129: EPIC data. That re-analysis resolved additional absorption lines in the

130: intermediate ($\sim$1--4~keV) energy band (Pounds and Page 2006),

131: strengthening the evidence for a high velocity outflow.  In total, 7

132: statistically significant absorption lines are identified, with the

133: strong $\sim$7 keV line re-asigned to He$\alpha$ of FeXXV, yielding an

134: {\it increased} outflow velocity  v$\sim$0.14$\pm$0.01c.  

135: 

136: 

137: Assuming a spherically symmetric flow, the mass loss rate is of then of

138: order  $\dot M \sim 35b\msun$~yr$^{-1}$, where $b\leq$1 allows for the

139: collimation of the flow. Modelling of the broad-band spectrum of

140: PG1211+143 (Pounds and Reeves 2006) has quantified both the absorbed and

141: re-emitted fluxes for the ionised outflow, yielding a covering factor

142: CF$\sim$0.1 for the high velocity,  highly ionised outflow, and $\dot M$

143: $ \sim 3.5\msun$~yr$^{-1}$. This outflow rate compares with $\dot M_{\rm

144: Edd}$ = 1.6$\msun$~yr$^{-1}$ for a non-rotating SMBH of mass

145: $\sim$$4\times 10{^7}$$\msun$ (Kaspi \etal\ 2000) accreting at an

146: efficiency of 10\%.  The mechanical energy of the highly ionised outflow

147: is then $\sim10^{45}$ erg s$^{-1}$, easily sufficient to power a

148: substantial  component of the X-ray emission of PG1211+143.

149: 

150: 

151: \section{A second continuum component}

152: For most AGN the energetically dominant emission component can be

153: modelled by a power law of photon index $\Gamma\sim$1.9 (Nandra and

154: Pounds 1994),  visible in Type 1 objects over the $\sim$2--100 keV band.

155: This `primary' continuum component is widely believed to arise by

156: Comptonisation of UV and optical accretion disc photons in a high

157: temperature electron corona (Haardt and Marashi 1991). However, recent

158: studies of spectral variability in MCG--6-30-15 (Vaughan and Fabian 2004)

159: and 1H0419-57 (Pounds \etal\ 2004) have provided clear evidence of a

160: variable power law component of slope significantly  steeper than the

161: `canonical value' of $\Gamma\sim$1.9. 

162: 

163: We now suggest this softer continuum component is physically distinct

164: from the `primary' continuum in PG1211+143, having established that the

165: mechanical energy in the fast, highly ionised outflow  is ample to power

166: that additional continuum component, perhaps via internal shocks 

167: (analogous to the process suggested in Gamma Ray Bursts), or shock

168: heating of slower moving clouds providing the bulk of the continuum

169: opacity.   The most likely X-ray emission mechanism would again be

170: Comptonisation of optical-UV disc photons, with the steep power law

171: resulting from a lower equilibrium temperature in the dispersed `corona'

172: compared with that responsible for the primary power law continuum.

173: 

174: A separate continuum component is attractive in the present context

175: since, if absorption is to create the impression of a strong `soft

176: excess', then the underlying continuum must be steeper than observed

177: above $\sim$2 keV. 

178: 

179: \begin{figure}

180: \rotatebox{-90}

181: {\includegraphics[width=.18\textheight]{praguefig3.ps}}

182: \rotatebox{-90}

183: {\includegraphics[width=.18\textheight]{praguefig4.ps}}

184: \caption{Left panel: deconvolution of the EPIC MOS spectrum of

185: PG1211+143 with the primary (dark grey) and secondary (light grey)

186: continua plus line emission from the  highly and moderately ionised

187: outflow (Pounds and Reeves 2006).  Right panel: structure in the soft

188: excess in the 2001 RGS data. XSTAR model fitting  shows this structure

189: is dominated by velocity-broadened resonance emission  lines of N,O,Ne

190: and Mg, together with a group of Fe-L lines at $\sim$0.7 keV.}

191: \label{3}

192: \end{figure}

193: 

194: 

195: \section{A new spectral model}

196: The evidence of absorption in highly ionised (narrow absorption lines)

197: and moderately ionised gas (low energy spectral curvature) suggests  a

198: new spectral model including 2 power law continua, both modified by

199: absorption, and re-emission from a 2-component photoionised gas.

200: Applying this model to the 2001 data from PG1211+143 produced a

201: statistically excellent fit (Pounds and Reeves 2006). 

202: 

203: Figure 2L illustrates the model components, showing that above $\sim$2

204: keV the spectrum is dominated by the `primary' power law, with a photon

205: index $\Gamma\sim$2.2, at the upper end of -- but consistent with -- the

206: `accepted' range for type 1 AGN spectra. At lower energies the 

207: secondary power law component ($\Gamma\sim$3.1) dominates, while below

208: $\sim$1 keV re-emission, particularly from the moderately ionised

209: absorber, becomes significant. 

210: 

211: Strong absorption of the primary power law is seen to be responsible for

212: the mid-band ($\sim$2--6 keV) spectral curvature, with ionised gas column

213: densities of N$_{H}\sim3-5\times 10^{22}$ cm$^{-2}$. In both pn and

214: MOS spectral fits the model shows a broad trough near $\sim$0.75--0.8

215: keV, due to Fe-L absorption. Of particular relevance in the present

216: context,  the luminosity of the steep, less absorbed, power law

217: component was $\sim6\times 10^{43}$~erg s$^{-1}$, more than an order of

218: magnitude lower than the estimated fast outflow energy.

219: 

220: In modelling spectral structure in the higher resolution RGS data

221: (Figure 2R),  photoionised line emission had to be convolved with a

222: gaussian ($\sigma\sim$25 eV at 0.6 keV), corresponding to velocity

223: broadening of  29000 km s$^{-1}$ (FWHM),  consistent with  the high

224: velocity flow. 

225: 

226: 

227: Applying the 2001 model to the 2004 data, a good fit was obtained with

228: element abundances and ionisation parameters fixed, the spectral change

229: being largely modelled by an increase in the steep, weakly absorbed

230: continuum component. In contrast, a change in the covering factor

231: (modelled by a lower column) of ionised absorber on the primary power

232: law was not able -- alone -- to fit both data sets.

233: 

234: 

235: \section{Broader application of the new spectral model}

236: 

237: The Black Hole Winds model (King and Pounds 2003) provides a simple

238: physical basis whereby massive, high velocity outflows can be expected

239: in AGN accreting at or above the Eddington limit. In addition to

240: PG1211+143, this might apply to both PDS456 (Reeves \etal\ 2003) and to

241: the bright Seyfert 1 IC4329A, recently shown (Markowitz \etal\ 2006) to

242: exhibit a strongly blue-shifted Fe K absorption line indicating a highly

243: ionised outflow at v$\sim$0.1c. While relatively rare in the local 

244: universe such X-ray spectra could be common for luminous, higher

245: redshift AGN.  

246: 

247: If the outflow velocity in PG1211+143 equates to the escape velocity at

248: the launch radius R$_{launch}$ (from an optically thick photosphere or

249: radiatively  extended inner disc), v$\sim$0.14c corresponds to

250: R$_{launch}$ $\sim$ $50R_{\rm s}$  (where $R_{\rm s} = 2GM/c^2$ is the

251: Schwarzschild radius), or $3\times 10^{14}$~cm for PG1211+143.  The EPIC

252: data show significant flux variability in the harder (2--10 keV) band on

253: timescales of 2--3 hours (fig 1 in Pounds  \etal\ 2003)), compatible with

254: the above scale size relating to the primary (disc/corona) X-ray

255: emission region.

256:  

257: Observations of PG1211+143 over several days are needed to test the new

258: spectral model. The variability timescale of the secondary power law

259: will constrain the scale of the region where we predict the fast outflow

260: undergoes internal shocks. Detecting hard X-ray emission from PG1211+143

261: above $\sim$20 keV would also support the need for a continuum component

262: with photon index less steep than that in the single power law/

263: absorption models,   while deeper RGS spectra are needed to resolve the

264: broad emission line profiles indicated in the model fits. 

265: 

266: In summary, it seems clear that further studies of high accretion rate

267: AGN, such as PG1211+143, offer great potential for understanding

268: outflows and their potential effects on the intergalactic medium and

269: host galaxy growth. 

270: 

271: 

272: \begin{thebibliography}{}

273: 

274: 

275: \bibitem[]{ho05} Chartas G., Brandt W.~N., Gallagher S.~C., Garmire G.~P. 2002, ApJ, 579, 169

276: 

277: \bibitem[]{be05} Chartas G., Brandt W.~N., Gallagher S.~C. 2003, ApJ, 595, 85 

278: 

279: \bibitem[]{fox05a} Chevallier L., Collin S., Dumont A-M., Czerny B., Mouchet M., Goncalves A.~C., Goosmann R. 2006, A\&A, 449, 493 

280: 

281: \bibitem[]{min05} Crummy J., Fabian A.~C., Gallo L., Ross R.~R. 2006, MNRAS, 365, 1067

282: 

283: \bibitem[]{nb05} Elvis M., Giommi P., Wilkes B.~J., McDowell J. 1991, ApJ, 378, 537

284: 

285: \bibitem[]{ge05} Gierlinski M., Done C. 2004, MNRAS, 349, L7 

286: 

287: \bibitem[]{ha91} Haardt F., Maraschi L. 1991, ApJ, 350, L81

288: 

289: \bibitem[]{dlk05} Kaspi S., Smith P.~S., Netzer H., Maoz D., Jannuzi B.~T., Giveon U. 2000, ApJ, 533, 631

290: 

291: \bibitem[]{ka06} Kaspi S., Behar E. 2006, ApJ, 636, 674

292: 

293: \bibitem[]{pa05a} King A.~R., Pounds K.~A. 2003, MNRAS, 345, 657

294: 

295: \bibitem[]{fox05b} King A.~R. 2005, ApJ, 635, L121

296: 

297: \bibitem[]{ma06} Markowitz A., Reeves J.~N., Braito V. 2006, ApJ, 646, 783

298: 

299: \bibitem[]{na94} Nandra K., Pounds K.~A. 1994, MNRAS, 268, 405

300: 

301: \bibitem[]{zm04} Pounds K.~A., Reeves J.~N., King A.~R., Page K.~L., O'Brien P.~T., Turner M.~J.~L. 2003, MNRAS, 345, 705

302: 

303: \bibitem[]{po04} Pounds K.~A., Reeves J.~N., Page K.~L., O'Brien P.~T. 2004, ApJ, 616, 696

304: 

305: \bibitem[]{laz01} Pounds K.~A., Page K.~L. 2006, MNRAS, 372, 1275

306: 

307: \bibitem[]{con} Pounds K.~A., Reeves J.~N. 2006, MNRAS, accepted (astro-ph/0610112)

308: 

309: \bibitem[]{p04} Proga D., Stone J.~M., Kallman T.~R. 2000, ApJ, 543, 686 

310: 

311: \bibitem[]{g05} Reeves J.~N., O'Brien P.~T., Ward M.~J. 2003, ApJ, 593, L65

312: 

313: \bibitem[]{n05} Schurch N.~J., Done C. 2006, MNRAS, 371, 81 

314: 

315: \bibitem[]{va04} Vaughan S., Fabian A.~C. 2004, MNRAS, 348, 1415

316: 

317: \end{thebibliography}

318: \bigskip

319: 

320: \discuss{Kazuo Makishima}{Comment: Steep power law you are finding is

321: extremely interesting and reminds me of similar phenomenon that occurs

322: in Galactic black hole binaries when they get in the very high state. In

323: that case this component appears between $10$--$50$~keV, connecting the

324: hard tail and the soft disc component.}

325: 

326: \discuss{Giorgio Matt}{Can you exclude that the $0.7$--$2$~keV spectrum is

327: due to ionized reflection rather than absorption?}

328: 

329: \discuss{Kenneth Pounds}{The spectral curvature (soft excess) in

330: PG1211 is too strong to be produced by photoionized reflection from the

331: accretion disc unless the direct continuum is somehow suppressed, for

332: example by light bending in strong gravity near the black hole. Given

333: the detection of a large column density outflow in this luminous AGN,

334: absorption plus re-emission provides a more likely explanation.}

335: 

336: \end{document}

337: