1: \documentclass{emulateapj}

2: \usepackage{apjfonts}

3: %%%\documentclass[12pt,preprint]{aastex}

4: %\usepackage{epsfig}

5: %\usepackage{lscape}

6: %\usepackage{gensymb}

7: 

8: \newcommand{\etal}{et al.~}

9: \def\gsim{\lower 2pt \hbox{$\, \buildrel {\scriptstyle >}\over

10: {\scriptstyle \sim}\,$}}

11: \def\lsim{\lower 2pt \hbox{$\, \buildrel {\scriptstyle <}\over

12: {\scriptstyle \sim}\,$}}

13: \def\rosat{{\sl ROSAT}}

14: \def\ciao{{\sl CIAO~}}

15: \def\ROSAT{{\sl ROSAT}}

16: \def\fuse{{\sl FUSE}}

17: \def\asca{{\sl ASCA}}

18: \def\ASCA{{\sl ASCA}}

19: \def\chandra{{\sl Chandra~}}

20: \def\oix{O~{\scriptsize IX}}

21: \def\oviii{O~{\scriptsize VIII}}

22: \def\ovii{O~{\scriptsize VII}}

23: \def\ovi{O~{\scriptsize VI}}

24: \def\oi{O~{\scriptsize I}}

25: \def\oii{O~{\scriptsize II}}

26: \def\oiv{O~{\scriptsize IV}}

27: \def\oiii{O~{\scriptsize III}}

28: \def\neviii{Ne~{\scriptsize VIII}}

29: \def\neix{Ne~{\scriptsize IX}}

30: \def\nex{Ne~{\scriptsize X}}

31: \def\HI{H~{\scriptsize I}\ }

32: \def\hii{H~{\scriptsize II}\ }

33: \def\aliii{Al~{\scriptsize III} }

34: \def\alii{Al~{\scriptsize II} }

35: \def\feii{Fe~{\scriptsize II}}

36: \def\fexvi{Fe~{\scriptsize XVI}}

37: \def\fexvii{Fe~{\scriptsize XVII}}

38: \def\fexviii{Fe~{\scriptsize XVIII}}

39: \newcommand{\as}{$^{\prime\prime}~$}

40: \newcommand{\am}{$^{\prime}~$}

41: \def\xs{GCSXE}

42: 

43: \shortauthors{Yao \& Wang}

44: \shorttitle{Galactic Central Diffuse X-Ray Enhancement}

45: 

46: \begin{document}

47: 

48: \slugcomment{\em Accepted for publication in the  Astrophysical Journal}

49: 

50: 

51: \title{The Galactic Central Diffuse X-Ray Enhancement: A Differential Absorption/Emission Analysis}

52: \author{Yangsen Yao\altaffilmark{1} and Q. Daniel. Wang\altaffilmark{2}}

53: \altaffiltext{1}{Massachusetts Institute of Technology (MIT) Kavli Institute 

54:  for Astrophysics and Space Research, 70 Vassar Street, Cambridge, MA 02139; 

55:  yaoys@space.mit.edu}

56: \altaffiltext{2}{Department of Astronomy, University of Massachusetts, 

57:   Amherst, MA 01003; wqd@astro.umass.edu}

58: 

59: \begin{abstract}

60: 

61: The soft X-ray background shows a general enhancement

62: toward the inner region of the Galaxy. But whether this enhancement is

63: a local feature (e.g., a superbubble within a distance of $\lsim 200$ pc) 

64: and/or a phenomenon related to energetic outflows from the Galactic center/bulge 

65: remains unclear. Here we report a comparative X-ray emission and 

66: absorption study of diffuse hot gas along the sight 

67: lines toward 3C~273 and Mrk~421, on and off the enhancement, but at similar

68: Galactic latitudes. The diffuse 3/4-keV emission intensity, as estimated from

69: the \rosat\ All Sky Survey, is about 3 times higher toward 3C~273 

70: than toward Mrk~421.  

71: Based on archival \chandra\ grating  observations of these two AGNs, 

72: we detect X-ray absorption lines (e.g., \ovii\ K$\alpha$, K$\beta$,

73: and \oviii\ K$\alpha$ transitions at $z \sim 0$) and find that the mean hot gas

74: thermal and kinematic properties along the two sight lines are significantly 

75: different. By subtracting the foreground and background contribution, 

76: as determined along the Mrk~421 sight line, 

77: we isolate the net X-ray absorption and emission produced by the hot gas 

78: associated with the enhancement in the direction of 3C~273. 

79: From a joint analysis of these differential data sets, we obtain the 

80: temperature, dispersion velocity, and hydrogen column density 

81: as $2.0(1.6, 2.3)\times10^6$ K, 216(104, 480)~km~s$^{-1}$, and

82: $2.2(1.4, 4.1)\times10^{19}~{\rm cm^{-2}}$, respectively (90\% confidence

83: intervals), assuming that the gas is approximately isothermal, solar

84: in metal abundances, and  equilibrium in collisional ionization. 

85: We also constrain the effective line-of-sight extent of the gas to be 

86: 3.4(1.0, 10.1) kpc, strongly suggesting that the enhancement 

87: most likely represents a Galactic central phenomenon.

88: 

89: \end{abstract}

90: 

91: \keywords{Galaxy: structure --- X-rays: ISM --- X-rays: individual (3C~273, Mrk 421)}

92: 

93: \section{Introduction }

94: \label{sec:intro}

95: The soft X-ray background (SXB) is greatly enhanced toward the inner part 

96: of the Galaxy ($l \lesssim 60^\circ$), particularly obvious in the 

97: \rosat\ All-Sky Survey (RASS) 

98: 3/4-keV band map (Fig.~\ref{fig:RASS}; Snowden et al. 1997). 

99: This Galactic central soft X-ray 

100: enhancement (GCSXE) apparently arises from diffuse hot 

101: gas; but its three-dimensional (3-D) morphology and origin are largely uncertain.

102: The most prominent part of the \xs\  is along a ridge that is commonly 

103: considered to be associated with the North Polar Spur (NPS) --- 

104: probably part of the radio continuum feature Loop I. 

105: The NPS is believed to be a local feature, based primarily on

106: stellar polarization measurements (Bingham 1967); but this assertion is

107: not universally accepted. The NPS and the rest of Loop I 

108: have been interpreted as a nearby ($D \sim 170$ pc) supernova remnant 

109: (SNR; e.g., Iwan 1980), a stellar wind bubble from the Scorpio-Centaurus 

110: OB association, or a combination of the two (e.g., Egger \& Aschenbach 1995).

111: Alternatively, the \xs\  could represent a recent outflow (e.g., a bipolar 

112: wind) or a giant explosion from the Galactic nuclear region (Sofue 1984; 

113: Bland-Hawthorn \& Cohen 2003), or a more gentle but lasting Galactic bulge 

114: wind interacting with the Galactic gaseous halo (\S~\ref{sec:dis}).

115: In this case, the physical size of the enhancement must be

116: comparable to our distance to the Galactic center ($\sim 8$ kpc).

117: Clearly, these scenarios have vastly different implications for

118: the overall energetics of the \xs\ (i.e., $10^{51}$ vs. 

119: $\gtrsim 10^{54}$ ergs) and its impact on the global 

120: interstellar medium (ISM) structure of 

121: the Galaxy. 

122: 

123: Studies of the \xs\ have been based primarily on broadband X-ray 

124: observations such as the RASS (Snowden et al. 1997) and more recently on

125: X-ray CCD observations from {\sl XMM-Newton} and {\sl Suzaku}. 

126: By observing the X-ray intensity variation on and off cool gas clouds 

127: at known distances, one can separate the foreground and background 

128: contributions (Burrows \& Mendenhall 1991; Wang \& Yu 1995; Park \etal 1997;

129: Almy \etal 2000; Kuntz \& Snowden 2000; Galeazzi \etal 

130: 2007; Smith \etal 2006; Henley \etal 2007). These experiments are particularly

131: useful in isolating the local X-ray emission and hence determining the 

132: properties of hot gas in the solar neighborhood. However, to infer

133: the properties of hot gas in distant regions is a much bigger challenge.

134: Particularly uncertain is the correction for the absorption by

135: intervening cool gas; the distribution of which relative to the X-ray-emitting

136: hot gas can be a complex (e.g, Kuntz \& Snowden 2000; 

137: Breitschwerdt \& de Avillez 2006). 

138: Nevertheless, it has been shown that

139: a large amounts of hot gas is present in regions toward the Galactic inner

140: region and beyond $d \sim 2 $ kpc (Almy et al. 2000). Along such a line of 

141: sight, one in general expects X-ray emission 

142: contributions from multiple components: 

143: the Local Hot Bubble (LHB; e.g., Snowden \etal 1998), 

144: the extended hot Galactic disk/corona 

145: (e.g., Yao \& Wang 2005, 2007), the 

146: unresolved Galactic and extragalactic point sources 

147: (e.g., Kuntz \& Snowden 2001; Hickox \& Markevitch 2006),

148: and the \xs. Of course, there could be additional discrete features

149: such as the NPS, which is most likely a separate identity, independent of

150: the \xs\ (see \S~4). Direct spectral decomposition of these components

151: along a single sight line is very difficult, if not impossible.

152: 

153: In this work, we present the first X-ray absorption line 

154: spectroscopy of the \xs, based on grating observations from

155: the {\sl Chandra X-Ray Observatory}. We compare the observations of 3C~273 and 

156: Mrk 421, on and off the \xs\ (Fig.~\ref{fig:RASS})

157: to determine its \ovii\ and \oviii\ line absorptions. This

158: differential spectroscopic data set enables us to characterize the thermal and 

159: kinematic properties of the hot gas associated with the \xs\ with

160: minimal uncertainties. The spectroscopic data, together with the 

161: emission intensity difference measured with the RASS, further allow us to

162: estimate the effective path-length and density of the 

163: hot gas.

164: 

165: \begin{figure}

166: \centerline{

167: %%%\includegraphics[width=0.45\textwidth]{figure/map.eps}}

168: \plotone{f1.eps}}

169: \caption{%% Directions of Mrk~421, 3C~273, LMC~X--3, 4U~1820--303, NPS,

170:   Directions of various relevant point sources (for X-ray absorption

171:  measurements) as well as the diffuse features: NPS and the radio loops I/IV 

172: (Haslam \etal 1971), shown on the RASS 3/4 keV band intensity 

173:   map \citep{sno97} in Aitoff projection.

174:   \label{fig:RASS} }

175: \end{figure}

176: 

177: 

178: Throughout the paper, we quote the errors at the 90\% confidence level, adopt the solar metal abundance from \citet{and89},

179: and assume that the hot gas is in the collisional ionization

180: equilibrium (CIE) state and is approximately isothermal. 

181: 

182: \section{Observations and Data Reduction}

183: 

184: As shown in Figure~\ref{fig:RASS}, the sight line of 3C~273 passes 

185: through the \xs, and is at the edges of radio loops I/IV. 

186: The 3/4-keV (RASS R4+R5 band) background is known to have quite 

187: a uniform intensity distribution, except for those abnormal regions 

188: such as the Galactic Bulge, NPS, the Cygnus Loop, etc. \citep{sno97}.

189: Much of the background emission in this band is proved to be thermal,

190: with distinct line emission in high spectral resolution observations

191: \citep{mcc02}. In contrast, the 1/4-keV (R1+R2 band) background intensity 

192: traces mostly hot gas in the LHB and residual emission 

193: from more distant regions after significant photo-electric absorption by 

194: foreground cool ISM. The background intensity in the 1/4-keV band

195: appears patchy and shows little correlation with those in the

196: higher energy bands. The background intensity in the 1.5-keV (R6+R7) band 

197: arises primarily from extragalactic AGNs. Therefore we only use the data in

198: 3/4-keV band that are primarily due to \ovii\ and \oviii\ line emission

199: in the hot ISM \citep{mcc02}.

200: 

201: The diffuse 3/4-keV background intensity

202: in the vicinity of the 3C~273 sight line is about 3 times of those at 

203: a typical high ($|b|\gsim20^\circ$) Galactic latitudes but away from 

204: the inner part of the Galaxy (e.g., Mrk~421 direction)

205: and is about 1/5 of that at the brightest part

206: of the NPS (Table~\ref{tab:ROSAT}; Snowden \etal 1997).

207: We use Mrk~421 direction as a reference sight line, which has a 

208: Galactic latitude similar to the 3C~273 direction. 

209: 

210: 

211: \begin{deluxetable}{lcccc}[b]

212: \tablewidth{0pt}

213: %%\tabletypesize{\footnotesize}

214: \tablecaption{Properties of the interested fields\label{tab:ROSAT}}

215: \tablehead{

216: %%          &   & $N_{\rm HI}$& SXB Int. & Diff. Int. \\

217: %%          & ($l, b$)   &(${\rm cm^{-2}}$)& (${\rm cts~s^{-1}~arcmin^{-2}}$)& (${\rm cts~s^{-1}~arcmin^{-2}}$) }

218:           &   & $N_{\rm HI}$ \\

219: objects          & ($l, b$)   &(${\rm cm^{-2}}$)& SXB Int. & Diffuse Int. }

220: \startdata

221:   3C~273 &($289\fdg95, ~64\fdg36$)&$1.7\times10^{20}$ &$1.92\times10^{-4}$ & $1.29\times10^{-4}$ \\

222:   Mrk~421&($179\fdg83, ~65\fdg03$)&$1.4\times10^{20}$ &$1.01\times10^{-4}$ & $0.39\times10^{-4}$ \\

223:   NPS    &($~26\fdg84, ~21\fdg96$)&$5.6\times10^{20}$ &$7.58\times10^{-4}$ & $~~6.95\times10^{-4}$ 

224: \enddata

225: \tablecomments{

226: %%   The intensities are in unit of ${\rm cts~s^{-1}~arcmin^{-2}}$. 

227: The total HI absorption column density are 

228:    adopted from \citet{sem01} for 3C~273 and from Dickey \& Lockman (1990) 

229:    for Mrk~421 and NPS directions. 

230:    The SXB intensities (in units of ${\rm counts~s^{-1}~arcmin^{-2}}$)

231:    toward 3C~273 and Mrk~421 are obtained from an annulus of $1\fdg0$ 

232:    to $1\fdg5$ radii around the source on the 3/4-keV band RASS

233:    map. The diffuse intensities are obtained by subtracting the 

234:    contributions from the unresolved extragalactic and Galactic point sources

235:    (\S~\ref{sec:intro}). The intensities of the NPS (its brightest part) 

236:    are obtained from a circular region of $1^\circ$  radius.

237: }

238: \end{deluxetable}

239: 

240: 

241: 

242: We use the RASS intensities in the R4 and R5 sub-bands to gain crude emission 

243: spectral characteristics of the 3C~273 and Mrk~421 sight lines. 

244: To account for the possible calibration uncertainties in the intensities

245: of the individual energy bands, we add 10\% 

246: \footnote{

247: The calibration uncertainty in intensities of the individual bands

248: is not available in the literature. The 10\% is our attempted value.}

249: systematic uncertainty to the emission spectrum in our data analysis.

250: To avoid the potential confusion from the 

251: emission of these bright sources due to the broad wing of the 

252: point spread function (PSF) of the RASS 

253: \footnote{For a PSPC pointing observation with an off-axis of $0'(30')$, the 

254: PSF density at 0.5 keV drops $\sim10^8(10^6)$ times at $\sim1^\circ$ away from 

255: the point source. For Mrk~421 at its historical high 

256: flux ($\sim50$ times higher than the normal flux; e.g., the {\sl Chandra} 

257: observation with ObsID 4148), the stray light from the point source 

258: contributes $\leq 0.2\%(20\%)$ to the SXB intensity of an annulus of 

259: $1\fdg0$ to $1\fdg5$ radii for R4+R5 band. The contribution from 3C~273 

260: to the SXB intensity is negligible.}

261: and to smooth out any potential local 

262: emission gradient, we extract the background intensities 

263: (Fig.~\ref{fig:ROSAT}) from an 

264: annulus of $1\fdg0$ to $1\fdg5$ radii centered at each of the sources, 

265: using the X-Ray Background Tool in HEASAC. 

266: \footnote{http://heasarc.gsfc.nasa.gov/cgi-bin/Tools/xraybg/xraybg.pl} 

267: 

268: 

269: 

270: \begin{figure}

271: \plotone{f2.eps}

272: \caption{Spectrum of the emission associated 

273: 	with the \xs\  and in the direction of 3C~273 ({\sl solid crosses}), 

274: 	the best-fit model 

275: 	convolved with the RASS response ({\sl solid histogram}), 

276: 	and the SXB spectra toward 3C~273 ({\sl dashed crosses}),

277: 	and Mrk~421 ({\sl dotted crosses}). See text for details.

278:    \label{fig:ROSAT} }

279: \end{figure}

280: 

281: 

282: 

283: The absorption spectra are extracted from an extensive set of 

284: {\sl Chandra} grating observations, with an accumulated exposure time 

285: of $\sim480$ ks each for 3C~273 and Mrk~421. We have

286: analyzed most of the observations of Mrk~421 in \citet{yao07} and have 

287: detected significant \ovii\ K$\alpha$, K$\beta$, and \ovii\ K$\alpha$ 

288: absorption lines in a co-added spectrum. 

289: For this study of the 3C~273 sight line, we use 13 observations taken 

290: with the Advanced CCD Imaging Spectrometer (ACIS). Among these

291: observations (ObsID: 1198, 2464, 

292: 2471, 3574, 4431, 5170, 459, 2463, 3456, 3457, 3573, 4430, 

293: and 5169), the last seven used the High Energy Transmission Grating 

294: (HETG), for which only the data from the medium-energy-grating arms are 

295: included. We do not use two observations taken with the High Resolution 

296: Camera for a total exposure of $\sim70$ ks; their inclusion would

297: increase the counting statistics only by $\sim15\%$ at $\sim$ 20 \AA, but 

298: would also introduce the complex overlapping from high ($>1$) order data

299: (Yao \& Wang 2006, 2007). We process the ACIS data, following the same 

300: procedure as described in Yao \& Wang (2006, 2007).

301: 

302: We focus on the oxygen absorption lines to avoid the uncertainty in 

303: dealing with relative metal abundances.

304: We fit the spectral continuum in the 18--22 \AA\ range

305: with a power law and fix the foreground cool gas absorption to be 

306: $N_H^c=1.7\times10^{20}~{\rm cm^{-2}}$

307: \citep{sem01}. This model fit gives $\chi^2$/dof = 524/397; the 

308: strong \ovii\ K$\alpha$, K$\beta$, and \oviii\ K$\alpha$

309: absorption lines are clearly visible in the residual.  

310: The inclusions of these absorption lines as described in

311: \S~3 significantly

312: improves the fit ($\chi^2/$dof $\simeq 402/392$). The resultant power-law 

313: index and the normalization are $2.1(2.0, 2.3)$ and

314: $2.3(2.2, 2.5)\times10^{-2}~{\rm photons~keV^{-1}~cm^{-2}~s^{-1}}$ 

315: at 1 keV. 

316: 

317: The \ovii\ K$\alpha$ absorption line  toward

318: 3C~273 has already been detected by \citet{fang03}, based on a subset of 

319: the observations used in this work. They have further discussed possible 

320: origins of the absorbing hot gas as a single component, either the 

321: intergalactic medium, the Galactic halo, or the radio loops I/IV. 

322: The detection of the additional absorption lines and the

323: comparative analysis between the sight lines, as reported 

324: here, now make it possible to isolate and characterize the contribution

325: from the \xs.

326: 

327: \section{Analysis and Results}

328: \label{sec:results}

329: We first characterize the absorbing gas by using our absline 

330: model\footnote{See \citet{yao05} and \citet{wang05} for a detailed 

331: description of the model.} to jointly fit the \ovii\ and \oviii\ absorption 

332: lines in the spectra of 3C~273 and Mrk~421. We measure the dispersion 

333: velocity ($v_b$), temperature ($T$), and the effective hydrogen column 

334: density ($N_{\rm H}$) of the

335: gas along each sight line. Table~\ref{tab:abs} summarize the results,

336: including the inferred \ovi, \ovii, and \oviii\ column densities. 

337: This single-temperature modeling represents only a simplified 

338: characterization of the gas. For example, a joint analysis

339: of the emission and absorption data for Mrk 421 shows that the 

340: hot gas along the sight line is not isothermal

341: \citep{yao07}. We expect a more complicated temperature structure

342: along the 3C~273 sight line. Nevertheless, the simple characterization, 

343: insensitive to the 

344: specific physical model adopted, should be sufficient for the modeling of 

345: averaged hot gas properties along the two sight lines,  

346: and thus facilitates a differential analysis of the \xs. 

347: 

348: \begin{deluxetable*}{lcccccc}

349: %%%\begin{deluxetable}{lcccccc}

350: %%%\rotate

351: \tabletypesize{\small}

352: \tablewidth{0pt}

353: \tablecaption{

354:   Spectral fit Results\label{tab:abs} }

355: \tablehead{

356: &$v_b$         & T   & N$_{\rm H}$ & N$_{\rm OVII}$ &N$_{\rm OVIII}$ & N$_{\rm OVI}$\\

357: direction & (km s$^{-1}$) & ($10^6$ K) & ($10^{19}$ cm$^{-2}$)    & ($10^{16}$ cm$^{-2}$)     &($10^{15}$ cm$^{-2}$) & ($10^{13}$ cm$^{-2}$)}

358: \startdata

359: 3C~273 & 139(92, 263) & 1.7(1.4, 2.0) &  3.5(2.3, 5.6)  & 2.0(1.3, 3.6) & 8.0(3.1, 14.7)  &6.0(3.8, 11.0)\\

360: Mrk~421 & 64(48, 104) & 1.4(1.3, 1.6) &  1.4(1.0, 2.0)  & 1.0(0.7, 1.5) & 1.5(0.6, ~2.6) &3.2(2.1, ~5.0)\\

361: GCSXE$^a$ & 216(104, 480) & 2.0(1.6, 2.3) &  2.2(1.4, 4.1) & 1.0(0.6, 2.6) & 7.2(2.0, 13.3)  &3.2(1.8, ~8.2)

362: \enddata

363: \tablecomments{In our absorption line model, the transition oscillation 

364:   coefficient and the damping factor are adopted from \citet{ver96}

365:   for \ovii\ and \oviii, and from  \citet{mor03} for \ovi. 

366:   90\% confidence intervals given in parenthesis.\\

367:   $^a$ The \xs\ 

368:   contribution along the sight line toward 3C~273 direction. See text for 

369:   the details.  

370: }

371: \end{deluxetable*}

372: 

373: 

374: 

375: Comparing with the Mrk~421 sight line, the 

376: 3C~273 sight line is clearly affected by the hot gas associated with the \xs.

377: The model characterizations of the two sight lines

378: are substantially different, in terms of both thermal and kinematic properties 

379: as well as the total absorbing column densities (Table~\ref{tab:abs}).

380: We attribute this difference to the additional absorption produced by the 

381: \xs. We characterize this net \xs\ absorption by including a separate 

382: component in modeling the absorption along the 3C~273 sight line

383: (Fig.~\ref{fig:counts_MRK421}). 

384: This component is in addition to the reference contribution

385: assumed to be the same as that estimated from the Mrk~421 sight line

386: (Table~\ref{tab:abs}). Uncertainties in the reference contribution are neglected

387: because the spectral signal-to-noise ratio of Mrk~421

388: (e.g., Fig.~1 in Yao \& Wang 2007) is a factor of $\sim4.5$ 

389: higher than that of 3C~273.  The parameters of the \xs\ 

390: component are all free to be fitted. 

391: 

392: \begin{figure} 

393: \plotone{f3.eps}

394: \caption{Detected oxygen absorption lines in the {\sl Chandra}

395: 	spectrum of 3C~273. The models have been convolved with the 

396: 	instrumental responses. The dotted histograms

397: mark the amount of absorptions toward Mrk~421 direction.

398: The bin size is 10 m\AA.

399:    \label{fig:counts_MRK421} }

400: \end{figure}

401: 

402: 

403: 

404: Similarly, we subtract the emission spectrum toward Mrk~421 from that 

405: toward 3C~273 to obtain the net emission spectrum of the \xs\ 

406: (Fig.~\ref{fig:ROSAT}). We have ignored the small difference in 

407: the foreground cool gas absorptions 

408: between the two sight lines (Table~\ref{tab:ROSAT}).

409: A correction for this difference will decrease

410: the emission intensity toward Mrk~421 by $\lsim5\%$, which is much less than

411: the statistical uncertainty (Fig.~\ref{fig:ROSAT}).

412: 

413: We measure the properties of the hot gas associated with the 

414: X-ray enhancement toward 3C~273 by jointly fitting the net emission 

415: and absorption spectra. This joint fit, as good as that for the absorption 

416: data alone, matches the emission spectrum well 

417: (Figs.~\ref{fig:ROSAT}-\ref{fig:counts_MRK421}).

418: The fit gives $v_b$, $T$, and $N_{\rm H}$ and further allows us to infer 

419: $N_{\rm OVII}$, $N_{\rm OVIII}$, and $N_{\rm OVI}$ for the hot 

420: gas associated with the \xs\ (Table~\ref{tab:abs}). The same fit also

421: gives an estimate of the effective path-length through the gas as 

422: $L=3.4(1.0, 10.1)$ kpc. This inference uses the fact that the 

423: emission measure (EM) of the hot gas is proportional to its number density 

424: square ($EM\propto A_{\rm O} n_H^2L\eta$), whereas 

425: the line absorption depends on the 

426: ionic column density ($N_H\propto N_{\rm OVII}/A_{\rm O} \propto n_HL\eta$). 

427: For simplicity,  both the oxygen

428: abundance $A_{\rm O}$ (in solar units) and the volume filling factor 

429: $\eta$ are set to 1 in the quoted results (see \S~\ref{sec:dis} 

430: for further discussions). 

431: Figure~\ref{fig:L_vs_T} shows the $L$ versus $T$ confidence contours.

432: With these characterizations, we further infer the average density,

433: pressure, and EM of the gas as 

434: $n_H = 2.1(0.4, 6.6)\times10^{-3}~{\rm cm^{-3}}$,

435: $P/k = 4.22(0.8, 13.3)\times10^3~{\rm cm^{-3}~K}$,

436: and $EM=1.9(0.6, 6.3)\times10^{-2}~{\rm cm^{-6}~pc}$.

437: 

438: \begin{figure} 

439: \plotone{f4.eps}

440: \caption{\small The 68\%, 90\%, and 99\% confidence contours of the

441: effective scale length ($L$) vs. the temperature ($T$) of the hot 

442: gas associated with the X-ray emission enhancement toward 3C~273.

443:    \label{fig:L_vs_T} }

444: \end{figure}

445: 

446: 

447: 

448: \section{Discussion}

449: \label{sec:dis}

450: 

451: We have obtained the dispersion velocity, temperature,

452: absorption column density, and the effective path-length of the diffuse gas 

453: associated with the X-ray emission enhancement toward the 3C~273 direction.

454: The 90\% lower limit of the path-length, $L\gsim1$ kpc, places the gas

455: well beyond any possible local origin within a distance of $D\lsim200$ kpc,

456: and therefore favors a remote origin of the gas.

457: The emission enhancement most likely reflects an outflow from the Galactic

458: center (GC) and/or bulge region. Large-scale winds driven by galactic nuclear 

459: starbursts and/or AGNs  have been observed 

460: in many galaxies (e.g., Martin 1999; Heckman \etal 2001). 

461: In our Galaxy, there is also evidence for the powerful energy injection 

462: from the GC on scales of several arcminutes to tens of degrees 

463: from radio to $\gamma$-ray 

464: (see Morris \& Serabyn 1996; Veilleux \etal 2005 for reviews).

465: The outflow is also expected from the energy and mass injection from

466: the Galactic bulge due to Type Ia supernovae and ejection of evolving 

467: low-mass stars. In any case, the outer boundary of the \xs\ may 

468: represent the interaction region of such an outflow with the 

469: Galactic disk and the accretion from the intergalactic medium

470: (Wang 2007). 

471: 

472: The large velocity dispersion ($v_b\sim216~{\rm km~s^{-1}}$) of the gas along

473: the 3C~273 sight line supports the GC origin of the \xs\ 

474: (Table~\ref{tab:abs}). Our estimate of the dispersion is based 

475: essentially on the relative saturation of the observed \ovii\ K$\alpha$

476: and K$\beta$ absorption lines and is consistent with the presence of

477: an \ovi\ high velocity absorption wing

478: (extending up to $\sim 240~{\rm km~s^{-1}}$), detected in the {\sl FUSE} 

479: spectrum of 3C~273 \citep{sem01}.

480: The dispersion is significantly higher than those estimated for the 

481: Mrk~421 sight line and for a similar off-\xs\ direction toward LMC~X--3 

482: (Galactic coordinates $l, b = 273\fdg57, -32\fdg08$; Fig.~\ref{fig:RASS};

483: Wang \etal 2005), indicating different natures of the intervening media.

484: %It is interesting to compare both the temperature and 

485: %large dispersion velocity of the hot gas toward 3C~273 with those

486: %toward 4U~1820-303 ($l, b = 2\fdg79, 7\fdg91$) 

487: %--- another on-\xs\ sight line (Fig.~\ref{fig:RASS}),

488: %close to the GC (e.g., Yao \& Wang 2006).

489: %The similar thermal and kinematic properties of the gas along these two

490: %sight lines suggest that they both trace the same phenomenon. 

491: At a temperature of $\sim2\times10^6$ K, the thermal broadening for the 

492: \ovii-bearing gas

493: is $\sim45~{\rm km~s^{-1}}$, which is negligible compared to 

494: $v_b$, indicating a large nonthermal broadening. 

495: This broadening cannot be due to the differential rotation of the Galactic 

496: disk, because of the high Galactic latitude direction and is most likely

497: due to the very turbulent nature of the gas, as is expected 

498: in the interaction region. 

499: 

500: It is also instructive to compare properties of the hot gas 

501: associated with the emission enhancement in the sight line of 3C~273 with 

502: those of the LHB. The temperature of the LHB is commonly accepted to be

503: $\sim10^6$ K, but the EM is still poorly constrained, 

504: ranging from 0.0079 to 0.02 ${\rm cm^{-6}~pc}$ (e.g., Smith \etal 2006; 

505: Galeazzi \etal 2007; Henley \etal 2007). Taking an average size of the LHB as 

506: 100 pc (e.g., Sfeir \etal 1999), the EM implies an absorbing 

507: column density of $2.7-4.4\times10^{18}~{\rm cm^{-2}}$ across the LHB. 

508: The obtained column density of the hot gas associated with the enhancement, 

509: $\sim2.2\times10^{19}~{\rm cm^{-2}}$ (Table~\ref{tab:abs}), can therefore 

510: accommodate 

511: $\sim$ 5-7 bubble boundaries like those of the LHB in the line of 

512: sight, implying a size of the hot gas $\gsim500$ pc. The temperature of the 

513: hot gas is $\sim2$ times higher than that of the LHB (Table~\ref{tab:abs}).

514: Therefore, if the thermal pressure could be assumed to be roughly the same 

515: as that in the LHB, the size of the \xs\ would then be larger. This simple 

516: comparison also suggests that the \xs\ represents a remote phenomenon.

517: 

518: How does the conclusion depend on our assumed solar metal abundances of

519: the hot gas? We have used only the X-ray absorption lines produced by 

520: oxygen ions. For a plasma at $\sim2\times10^6$ K, \ovii\ and \oviii\ 

521: dominates ($\geq85\%$) the emission in the 

522: RASS 3/4-keV band. Therefore, a deviation of non-oxygen metal 

523: elemental abundances from the assumed solar values will not significantly 

524: change our $L$ estimate. For example, assuming the abundances 

525: equal to zero would give $L=2.8(0.9, 9.1)$ kpc, whereas adopting the

526: abundances to be supers-solar would lead to a slightly larger $L$.

527: The $L$ is nearly inversely proportional to the oxygen abundance.

528: To let the \xs\ be consistent with a local

529: origin (i.e., $L\lsim200$ pc; \S~\ref{sec:intro}) the oxygen 

530: abundance needs to be $\gsim5$ (90\% lower limit; \S~\ref{sec:results})

531: times solar, which is very unlikely.

532: 

533: With the characterizations of the hot gas obtained in \S~\ref{sec:results}, 

534: we can estimate the thermal energy

535: contained in wind material. We take a simple bipolar glass-timer 

536: geometry for the wind with a cone-like shape

537: near the GC and a cylindric shape at large vertical distances

538: ($5\leq |z|\leq20$ kpc; e.g., Sofue 2000; Bland-Hawthorn \& Cohen 2003).

539: The radius of the cylinder is assumed to be $\sim7.2$ kpc, to be consistent

540: with our estimated path-length ($L \sim 3.4$ kpc) along the 3C~273 

541: sight line (taking the distance to the Galactic center of 7.6 kpc; 

542: Eisenhauer \etal 2005). To simplify our calculation, we assume 

543: that the gas density is uniformly distributed and the

544: gas temperature is exponentially decaying along the vertical direction from 

545: $5\times10^6$ K near the GC (e.g., Park \etal 1997) to $2\times10^6$ K at the 

546: latitude of 3C~273 (Table~\ref{tab:abs}). We find that total energy is 

547: $\sim10^{56}$ ergs, which is roughly in line with that expected in the

548: GC wind models \citep{sof00, bla03}.

549: 

550: While the emission enhancement toward 3C~273 likely arises in a region around

551: the GC, the bright NPS may still have a local origin 

552: (e.g., de Geus 1992). In this case, the NPS may be superposed on the 

553: large-scale \xs, which appears much more extended (to regions with $l \gtrsim

554: 60^\circ$; Fig.~1). \citet{mil06} recently presented a high resolution 

555: {\sl Suzaku} emission spectrum of the bright NPS (Table~\ref{tab:ROSAT}).

556: From a multiple component fit, they found that while the relative 

557: abundances of Ne/O, Fe/O, and Mg/O are consistent with the solar values, 

558: the N/O is $\sim4$ times higher. They therefore attributed the NPS to

559: the Sco-Cen environment, being enriched by material 

560: undergoing the CNO cycle in massive stars. In addition,

561: the temperature they obtained,  $\sim3.4\times10^6$ K,

562: is substantially higher than that of the \xs, as we have obtained here

563: (Table~\ref{tab:abs}). A similarly high temperature was also obtained 

564: from the {\sl XMM-Newton} observations \citep{wil03}. The apparent high 

565: temperature of the NPS may be due to the high-velocity shock-heating, 

566: commonly seen in young supernova remnants. 

567: 

568: We have assumed that the hot gas associated with the \xs\ along the

569: 3C~273 sight line is isothermal, which is adequate in the present study 

570: due to the very limited spectral resolution of the RASS data used here. 

571: Emission data with a low instrument background and relatively high

572: spectral resolution (e.g., from a deep {\sl Suzaku} X-ray CCD observation)

573: will be particularly useful for tightening the constraints on the properties of 

574: the hot gas associated with the \xs.

575: 

576: \acknowledgements 

577: We thank Eric Miller, Claude Canizares, Paola Testa, and 

578: Herman Marshall for useful discussions. We are also grateful to 

579: the anonymous referee 

580: for insightful comments and suggestions, which helped to 

581: improve the presentation of the paper.

582: This work is supported by NASA through the 

583: Smithsonian Astrophysical Observatory contract SV3-73016 to 

584: MIT for support of the {\sl Chandra} X-Ray Center under contract NAS 08-03060. 

585: Support from {\sl Chandra} archival research grant AR7-8016 are also 

586: acknowledged. QDW appreciates additional support from NASA grant NNX06AB99G.

587: 

588: 

589: \begin{thebibliography}{}

590: \bibitem[Almy \etal (2000)]{alm00} Almy, R. C., McCammon, D., Digel, S. W., \etal 2000, ApJ, 545, 290

591: \bibitem[Anders \& Grevese (1989)]{and89} Anders, E., \& Grevesse, N. 1989, Geochim. Cosmochim. Acta, 53, 197

592: \bibitem[Bingham (1967)]{bin67} Bingham, R. G. 1967, MNRAS, 137, 157

593: \bibitem[Bland-Hawthorn \& Cohen (2003)]{bla03}Bland-Hawthorn, J., \& Cohen, M. 2003, ApJ, 582, 246

594: \bibitem[Breitschwerdt \& de Avillez (2006)]{bre06} Breitschwerdt, D., \& de Avillez, M. A. 2006, A\&A, 452L, 1

595: \bibitem[Burrows \& Mendenhall (1991)]{bur91} Burrows, D. N., \& Mendenhall, J. A. 1991, Nature, 351, 629

596: %%\bibitem[Cheng \etal (1997)]{che97} Cheng, L. X., Leventhal, M., Smith, D. M., \etal 1997, \apj, 481, L43

597: %%\bibitem[Dalgarno \& Lallement (2004)]{dal04} Dalgarno, A., \& Lallement, R. 2004, \apj, 617, 1347

598: \bibitem[de Geus (1992)]{deg92} de Geus, E. J. 1992, A\&A, 262, 258

599: %%\bibitem[Dermer \& Skibo (1997)]{der97} Dermer, C. D., \& Skibo, J. G. 1997, \apj, 487, L57

600: \bibitem[Dickey \& Lockman (1990)]{dic90} Dickey, J. M., \& Lockman, F. J. 1990, \araa, 28, 215

601: \bibitem[Egger \& Aschenbach (1995)]{egg95} Egger, R. J., \& Aschenbach, B. 1995, A\&A, 294, L25

602: \bibitem[Eisenhauer \etal (2005)]{eis05} Eisenhauer, F., Genzel, R., Alexander, T., \etal 2005, \apj, 628, 246

603: \bibitem[Heckman \etal (2001)]{hec01} Heckman, T. M., Sembach, K. R., Meurer, G. R., \etal 2001, \apj, 554, 1021

604: \bibitem[Henley \etal (2007)]{hen07} Henley, D. B., Shelton, R. L., \& Kuntz, K. D. \apj\ 661, 304

605: \bibitem[Fang \etal (2003)]{fang03} Fang, T., Sembach, K., \& Canizares, C. 2003, \apjl, 586, L49

606: \bibitem[Galeazzi \etal (2007)]{gal06} Galeazzi, M., Gupta, A., Covey, K., \& Ursino, E. 2007, \apj, 658, 1081

607: \bibitem[Haslam \etal (1971)]{has71} Haslam, C. G. T., Kahn, F. D., \& Meaburn, J. 1971, A\&A, 12, 388

608: \bibitem[Hickox \& Markevitch (2006)]{hic06}Hickox, R., \& Markevitch, M. 2006, ApJ, 645, 95

609: \bibitem[Iwan (1980)]{iwa80} Iwan, D. 1980, ApJ, 239, 316

610: %%\bibitem[Keeney \etal (2006)]{kee06} Keeney, B. A., Danforth, C. W., Stocke, J. T., \etal, 2006, \apj, 646, 951

611: \bibitem[Kuntz \& Snowden (2000)]{kun00} Kuntz, K. D., \& Snowden, S. L. 2000, ApJ, 543, 195       

612: \bibitem[Kuntz \& Snowden (2001)]{kun01} ------. 2001, ApJ, 554, 684       

613: %%\bibitem[Lockman (1984)]{loc84} Lockman, F. J. 1984, \apj, 283, 90

614: \bibitem[Martin (1999)]{mar99} Martin, C. L. 1999, \apj, 513, 156

615: \bibitem[McCammon \etal (2002)]{mcc02} McCammon, D., Almy, R., Apodaca, E., \etal 2002, \apj, 576, 188

616: \bibitem[Miller \etal (2006)]{mil06} Miller, E., Tsunemi, H., Bautz, M., \etal 2006, AAS, 208, 3504

617: \bibitem[Morris \& Serabyn (1996)]{mor96}Morris, M., \& Serabyn E. 1996, \araa, 34, 645

618: \bibitem[Morton (2003)]{mor03} Morton, D. C. 2003, \apjs, 149, 205

619: %%\bibitem[Preibisch \& Zinnecker (1999)]{pre99} Preibisch, T., \& Zinnecker, H. 1999, \aj, 117, 2381

620: \bibitem[Park \etal (1997)]{park97} Park, S., Finley, J. P., Snowden, S. L., \& Dame, T. M. 1997, \apj, 476, L77

621: \bibitem[Sembach \etal (2001)]{sem01} Sembach, K. R., Howk, J. C., Savage, B. D., Shull, \etal 2001, \apj, 561, 573

622: %%\bibitem[Sembach \etal (2003)]{sem03}Sembach, K. R., Wakker, B. P., Savage, B. D., \etal 2003, ApJ S., 146, 165

623: \bibitem[Sfeir \etal (1999)]{sfe99} Sfeir, D. M., Lallement, R., Crifo, F., \& Welsh, B. Y. 1999, A\&A, 346, 785 

624: %\bibitem[Smith \etal (2005)]{smi05} Smith, R. K., Edgar, R. J., Plucinsky, P. P., \etal 2005, \apj, 623, 225

625: \bibitem[Smith \etal (2006)]{smi06} Smith, R. K., \etal 2006, \pasj\ 59, 141 

626: \bibitem[Snowden \etal (1997)]{sno97} Snowden, S. L., \etal 1997, \apj, 485. 125

627: \bibitem[Snowden \etal (1998)]{sno98} ------. 1998, ApJ, 493, 715

628: \bibitem[Sofue (1984)]{sof84} Sofue, Y. 1984, Nature, 310, 568

629: \bibitem[Sofue (2000)]{sof00} ------. 2000, \apj, 540, 224

630: \bibitem[Veilleux \etal (2005)]{vei05} Veilleux, S., Cecil, G., \& Bland-Hawthorn 2005, \araa, 43, 769

631: \bibitem[Verner \etal (1996)]{ver96} Verner, D, A., Verner, E. M., \& Ferland, G. J. 1996, Atomic Data \& Nuclear Data Tables, 64, 1-180

632: \bibitem[Wang \& Yu (1995)]{wang95} Wang, Q. D. \& Yu, K. 1995,  AJ, 102, 698

633: \bibitem[Wang \etal (2005)]{wang05} Wang, Q. D., Yao, Y., Tripp, T. M., \etal 2005, \apj, 635, 386

634: \bibitem[Wang (2006)]{wang06} Wang, Q. D. in the proceedings of Chemodynamics: from first stars to local galaxies, in press (astro-ph/0611038)

635: \bibitem[Willingale \etal (2003)]{wil03}Willingale, R., Hands, A. D. P., Warwick, R. S., \etal 2003, MNRAS, 343, 995

636: \bibitem[Yao \& Wang (2005)]{yao05} Yao, Y., \& Wang, Q. D. 2005, \apj, 624, 751 

637: \bibitem[Yao \& Wang (2006)]{yao06} ------. 2006, \apj, 641, 930

638: %\bibitem[Yao \etal (2006)]{yao06b} Yao, Y., Norbert, S., Wang, Q. D., \& Nowak, M. 2006, \apjl, 653, L121

639: \bibitem[Yao \& Wang (2007)]{yao07} ------. 2007, \apj, 658, 1088

640: %%\bibitem[Yusef-Zadeh \etal (2000)]{yus00} Yusef-Zadeh, F., Melia, F., \& Wardle, M. 2000, Science, 287, 85

641: \end{thebibliography}

642: 

643: \clearpage

644: 

645: 

646: \clearpage

647: 

648: 

649: \end{document}

650: 

651: