1: \documentclass{pasj00}
2: %\draft
3:
4: \SetRunningHead{Hyodo et al.}{Suzaku Spectroscopy of M17}
5:
6: %\Received{\today}
7: %\Received{//}
8: %\Accepted{//}
9:
10: \begin{document}
11: \title{Suzaku Spectroscopy of the Extended X-Ray Emission in M17}
12: %
13: \author{Yoshiaki~\textsc{Hyodo},\altaffilmark{1}
14: Masahiro~\textsc{Tsujimoto},\altaffilmark{2,3,4}
15: Kenji~\textsc{Hamaguchi},\altaffilmark{5,6} Katsuji~\textsc{Koyama},\altaffilmark{1}\\
16: Shunji~\textsc{Kitamoto},\altaffilmark{2} Yoshitomo~\textsc{Maeda},\altaffilmark{7}
17: Yohko~\textsc{Tsuboi}\altaffilmark{8}, and Yuichiro~\textsc{Ezoe}\altaffilmark{9}}
18: %
19: \altaffiltext{1}{Department of Physics, Graduate School of Science, Kyoto University,\\
20: Kita-shirakawa Oiwake-cho, Sakyo, Kyoto 606-8502}
21: \altaffiltext{2}{Department of Physics, Rikkyo University, 3-34-1 Nishi-Ikebukuro, Toshima, Tokyo 171-8501}
22: \altaffiltext{3}{Department of Astronomy \& Astrophysics, Pennsylvania State University,\\
23: 525 Davey Laboratory, University Park, PA 16802, USA}
24: \altaffiltext{4}{Chandra Fellow}
25: \altaffiltext{5}{CRESST and X-ray Astrophysics Laboratory, Goddard Space Flight Center,\\
26: National Aeronautics and Space Science, Greenbelt, MD 20771, USA}
27: \altaffiltext{6}{Universities Space Research Association, 10211 Wincopin Circle, Suite 500, Columbia, MD 21044, USA}
28: \altaffiltext{7}{Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency,\\
29: 3-1-1 Yoshinodai, Sagamihara, Kanagawa 229-8510}
30: \altaffiltext{8}{Department of Science and Engineering, Chuo University, 1-13-27 Kasuga, Bunkyo, Tokyo 112-8551}
31: \altaffiltext{9}{Cosmic Radiation Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198}
32:
33: %
34: \email{hyodo@cr.scphys.kyoto-u.ac.jp}
35: \KeyWords{X-rays: ISM --- ISM: bubbles --- ISM: H\emissiontype{II} regions --- Galaxy: open clusters and associations: individual (M17)}
36: \maketitle
37:
38: \begin{abstract}
39: We present the results of a Suzaku spectroscopic study of the soft extended X-ray
40: emission in the H\emissiontype{II} region M17. The spectrum of the extended emission
41: was obtained with a high signal-to-noise ratio in a spatially-resolved manner using the
42: X-ray Imaging Spectrometer (XIS). We established that the contamination by unresolved
43: point sources, the Galactic Ridge X-ray emission, the cosmic X-ray background, and the
44: local hot bubble emission is negligible in the background-subtracted XIS spectrum of
45: the diffuse emission. Half a dozen of emission lines were resolved clearly for the
46: first time, including K$\alpha$ lines of highly ionized O, Ne, and Mg as well as L
47: series complex of Fe at 0.5--1.5~keV. Based on the diagnosis of these lines, we
48: obtained the following results: (1) the extended emission is an optically-thin thermal
49: plasma represented well by a single temperature of $\sim$3.0\,$\pm$\,0.4~MK, (2) the
50: abundances of elements with emission lines in the diffuse spectrum are 0.1--0.3~solar,
51: while those of bright discrete sources are 0.3--1.5~solar, (3) the metal abundances
52: relative to each other in the diffuse emission are consistent with solar except for a
53: Ne enhancement of a factor of $\sim$2, (4) both the plasma temperature and the chemical
54: composition of the diffuse emission show no spatial variation across the studied
55: spatial scale of $\sim$5~pc.
56: \end{abstract}
57:
58:
59: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
60: \section{Introduction}
61: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
62: Massive stars are a driving force of physical and chemical evolutions of their host
63: galaxies. Supernova explosions and their remnants have been intensively studied for
64: decades, but the pre-explosion effects are equally important; the integrated mass,
65: momentum, and energy releases over the lifetime of an O star can be comparable to those
66: by a supernova explosion at the end of their lives \citep{leitherer92}. Diffuse X-ray
67: emission is generated as a consequence of shocks by stellar winds impinging on the
68: interstellar medium (ISM). Therefore, we can quantitatively study the effects of energy
69: dissipation and the chemical enrichment of interstellar space by early-type stars
70: through the spectroscopy of X-ray emission in H\emissiontype{II} regions.
71:
72: \citet{weaver77} presented a self-similar solution of stellar winds interacting with the
73: ISM. They showed a single O7 star forms a hot ($\sim$10$^{6-7}$~K) bubble by the shock,
74: which can be observed as soft extended ($\sim$10~pc) X-ray emission. \citet{townsley03}
75: claimed the first unambiguous detections of such emission in the H\emissiontype{II}
76: regions M17 and the Rosette Nebula using the Advanced CCD Imaging Spectrometer (ACIS;
77: \cite{garmire03}) onboard the Chandra X-ray Observatory \citep{weisskopf02}. With a
78: $\sim$40~ks integration time of M17 (ObsID\,$=$\,972), diffuse soft X-ray emission was
79: detected apart from 886 point sources above $\sim$10$^{29.3}$~erg~s$^{-1}$
80: \citep{broos07}. \citet{dunne03} showed the entire structure of the soft X-ray diffuse
81: emission using the Position-Sensitive Proportional Counter (PSPC; \cite{pfeffermann87})
82: onboard ROSAT \citep{trumper82}. They measured the total X-ray luminosity and compared
83: to the wind-blown bubble models with and without heat conduction. They concluded that
84: only the bubble without heat conduction can account for the observed X-ray
85: luminosity of $\sim 2.5\times10^{33}$~erg~s$^{-1}$. The magnetic field may be
86: responsible for suppressing the heat conduction and mass evaporation between the hot gas
87: and cold ISM \citep{dunne03}.
88:
89: The results obtained by these high-resolution imaging studies are generally consistent
90: with the wind-blown bubble models. However, the observational results are not still
91: accurate enough to compare to the theoretical works. It is often ambiguous whether the
92: observed diffuse emission is from wind-blown bubbles or from supernovae, which give rise
93: to diffuse emission with a similar X-ray spectral hardness and luminosity in a similar
94: spatial scale. The largest uncertainty stems from the lack of spectral analysis based on
95: line diagnostics in a spatially-resolved manner.
96:
97: Resolving emission lines is crucial to examine whether the spectrum is thermal and to
98: determine the temperature and the chemical composition of the plasma. For example, the
99: intensity ratio of K$\alpha$ lines between O\emissiontype{VII} and O\emissiontype{VIII}
100: and that between Ne\emissiontype{IX} and Ne\emissiontype{X} are steep functions of the
101: plasma temperature at 1--10~MK \citep{tucker66}. In between the O and Ne K$\alpha$
102: complex, Fe L series lines dominate the spectrum. The metallicity of these elements is
103: one of the factors to determine the X-ray luminosity expected from a wind-blown bubble
104: \citep{chu95}. The anomaly in the O and Fe abundance ratio can be used to discriminate
105: different types of supernovae \citep{tsujimoto95,nomoto97} if the emission is of a
106: supernova origin. The previous studies using ROSAT and Chandra were incapable of
107: resolving these lines, limiting their ability to diagnose the plasma emission.
108:
109: The X-ray Imaging Spectrometer (XIS; \cite{koyama07}) onboard Suzaku \citep{mitsuda07}
110: has a superior spectral resolution, a low background, and a large effective area, which
111: are particularly suited for spectroscopy of extended X-ray emission. The capability to
112: resolve key elements with sufficient statistical significance has been illustrated by
113: several initial studies on extended emission in H\emissiontype{II} regions
114: \citep{hamaguchi07,tsujimoto07}. \citet{hamaguchi07} resolved various emission lines
115: from the diffuse emission in the Carina Nebula. Based on the low nitrogen-to-oxygen
116: ratio and the spatial variation of the Fe and Si abundances, they suggested that the
117: diffuse emission originates not from wind-blown bubbles but from one or multiple old
118: supernova remnant(s).
119:
120: \medskip
121:
122: M17 is a Galactic H\emissiontype{II} region at a distance of $\sim$1.6~kpc
123: \citep{nielbock01}. \citet{hanson97} identified nine O stars in the central OB
124: association with near infrared spectroscopy. A much larger number of young OB stars are
125: suggested by near infrared photometry \citep{lada91,jiang02}. The earliest system is a
126: binary of two O4--O5 stars. The age of the cluster is estimated to be $\lesssim$1~Myr
127: based on the H-R diagram \citep{hanson97}. Strong winds and radiation from the central
128: OB association sculpted the ambient matter to form a \textsf{V}-shaped cloud, which was
129: traced by molecular and atomic hydrogen lines
130: \citep{chrysostomou02,felli84,brogan01}. The diffuse X-ray emission found by Chandra and
131: ROSAT has an asymmetric morphology with respect to the OB association and fills the
132: cavity of the molecular cloud toward the negative Galactic latitude \citep{povich07}.
133:
134: M17 is suitable for X-ray studies of hot bubbles in H\emissiontype{II} regions for being
135: proximate and having a very high contrast of the diffuse emission against point
136: sources. From the Chandra study \citep{townsley03}, the diffuse emission is more intense
137: than the integrated emission of resolved point sources by more than ten-fold at
138: 1~keV. Despite the limited spatial resolution of Suzaku, therefore, the obtained
139: spectrum is not seriously contaminated by unresolved point sources.
140:
141:
142: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
143: \section{Observation}
144: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
145: Suzaku observed M17 on 2006 March 11--14 in the first announcement of opportunity
146: observing cycle. Suzaku has XIS and Hard X-ray Detector (HXD;
147: \cite{kokubun07,takahashi07}) instruments. We concentrate on the XIS data in this paper,
148: which has a sensitivity for the soft emission studied here.
149:
150: The XIS is equipped with four X-ray CCDs. Three of them (XIS0, 2, and 3) are
151: front-illuminated (FI) CCDs and the remaining one (XIS1) is a back-illuminated (BI)
152: CCD. Each CCD chip has a format of 1024$\times$1024 pixels and is composed of four
153: segments of 256$\times$1024 pixels. FI and BI CCDs are superior to each other in the
154: hard and soft band responses, respectively. They are mounted at the focus of four
155: independent X-ray telescopes (XRT; \cite{serlemitsos07}). The detectors are sensitive in
156: the energy range of 0.2--12.0~keV with an initial energy resolution of $\sim$65~eV in
157: the full width at half maximum and a total effective area of $\sim$1360~cm$^{2}$ at
158: 1.5~keV. An XIS field of view covers a $\sim$18\arcmin $\times$18\arcmin\ region with a
159: half power diameter of $\sim$2\arcmin. The radioactive sources of \atom{Fe}{}{55}
160: illuminate two corners of each of the four CCDs for calibration purposes.
161:
162: The performance of XIS is subject to degradation due to the radiation damage in the
163: orbit \citep{koyama07}. As of the observation date, the energy resolution in the full
164: width at half maximum is $\sim$90~eV at 1.5~keV. The relative energy gains among XIS
165: chips and segments have a systematic uncertainty of $\sim$5~eV. An unknown contaminant
166: accumulates on the optical blocking filters of the XIS in the orbit. As a result, the
167: effective area at the soft band end has diminished significantly, making the carbon and
168: nitrogen features difficult to detect below $\sim$0.5~keV. This effect is included in
169: the auxiliary response.
170:
171: The observation was conducted using the normal clocking mode with a frame time of
172: 8~s. Data (revision 1.2\footnote{See http://www.astro.isas.jaxa.jp/suzaku/process/ for
173: details.}) were screened to remove events during the South Atlantic Anomaly passages, at
174: elevation angles below 4$^\circ$ from the earth rim, and at elevation angles below
175: 10$^\circ$ from sunlit earth rim. We constructed plots of the raw count rate versus the
176: elevation angles to find that these criteria maximize the exposure time with negligible
177: contaminating emission. After the filtering, the net integration time is $\sim$110~ks.
178:
179:
180:
181: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
182: \section{Analysis}
183: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
184: \subsection{Image Analysis}
185: \begin{figure*}
186: \begin{center}
187: \FigureFile(85mm,85mm){f1a.eps}
188: \FigureFile(85mm,85mm){f1b.eps}
189: \end{center}
190: %
191: \caption{XIS images in the (a) 0.5--1.5~keV and (b) 1.5--5~keV bands. The XIS field is
192: shown with the solid square in both panels, while the ACIS field is shown with the
193: dashed square in (b). The source and background regions for diffuse emission are shown
194: by solid lines in (a). The source sub-regions (a, b, and c) are ruled by dashed
195: curves. The discrete sources are shown with arrows in (b). Both images are processed as
196: follows: (1) Non--X-ray background constructed from night earth observations was
197: subtracted. (2) The astrometry and vignetting were corrected. (3) The images were
198: adaptively binned to achieve a signal-to-noise ratio of larger than 8 using the
199: weighted Voronoi tessellation algorithm (\cite{Diehl06},
200: \cite{cappellari03}).}\label{fg:f1}
201: %
202: \end{figure*}
203:
204: Figure~\ref{fg:f1} shows the XIS images of the study field in the (a) soft
205: (0.5--1.5~keV) and (b) hard (1.5--5.0~keV) bands. The two band-limited images appear
206: strikingly different. In the hard band, the image is dominated by the emission from the
207: OB association. We also see the excess emission from a group of protostars M17 North
208: \citep{wilson79,henning98,broos07}. In the soft band, the extended emission emerges in
209: the eastward of the OB association, as was claimed by \citet{townsley03} and
210: \citet{dunne03}. The XIS observation was centered at the most intense part of the
211: extended emission at (R.A., decl.) $\sim$ (\timeform{18h20m50s}, --\timeform{16D12'}) in
212: the equinox J2000.0, while the ACIS observation was at the OB association
213: \citep{townsley03}. The XIS and ACIS images have a similar size ($\sim$18 and $\sim$17
214: arc-minute square, respectively) with a $\sim$70\% overlapping area
215: (figure~\ref{fg:f1}b).
216:
217: The astrometry of the XIS frame was registered using 1WGA\,J1820.6--01615 found in both
218: images. The O8 star \citep{ogura76,white94} is bright, isolated, and point-like in the
219: ACIS image (CXOU\,J182035.87--161542.5; \cite{broos07}), thus serves as a good
220: astrometric calibrator. We shifted the XIS frame by $\sim$17\arcsec\ to the north so
221: that the position matches with that by the ACIS observation. The Chandra frame is
222: accurate to $\sim$0\farcs5 in the astrometry\footnote{See
223: http://cxc.harvard.edu/proposer/POG for details.}.
224:
225:
226: \subsection{Spectral Analysis of Extended Emission}
227: \subsubsection{Entire Emission}
228: We first examine the spectrum of the entire diffuse emission. We extracted the source
229: spectrum from a rectangular region and the background spectrum from a region devoid of
230: intense diffuse emission (figure~\ref{fg:f1}a). Because the off-axis angles of the
231: source and background regions are different, we processed the raw spectra in the
232: following way before subtracting the background from the source: (1) The
233: non--X-ray-background (NXB) spectrum was subtracted, which was constructed from night
234: earth data at the same extraction region. The NXB of XIS is a function of the
235: geomagnetic cut-off rigidity. We therefore compiled night earth observations such that the
236: cut-off rigidity distribution becomes the same with that of the M17 observation. (2) The
237: vignetting was corrected by multiplying the effective area ratios between the source and
238: background regions for each energy bin of the background spectrum. This takes into
239: account the accumulating contaminant on the XIS optical blocking filter.
240:
241: The merged FI and the BI spectra are shown in figure \ref{fg:f2}. In the merged FI spectrum,
242: we added the three FI spectra to increase the photon statistics, because the
243: redistribution matrix functions (RMFs) and the auxiliary response functions (ARFs) are
244: essentially the same for these chips. On the other hand, we handled the BI spectrum
245: separately for its different response.
246:
247: \begin{figure}[!ht]
248: \begin{center}
249: \FigureFile(85mm,85mm){f2.eps}
250: \end{center}
251: \caption{Background-subtracted XIS spectra of the entire diffuse emission (regions
252: a$+$b$+$c in figure~\ref{fg:f1}a). BI spectrum is shown in red, while the merged FI
253: spectrum is in black in a linear scale. Conspicuous emission lines are labeled, which are
254: K$\alpha$ lines except for the Fe L series lines.}\label{fg:f2}
255: \end{figure}
256:
257: We resolved the emission lines clearly for the first time, which include K shell lines
258: of O, Ne, and Mg as well as L shell lines of Fe. This indicates that the emission is of
259: a thermal origin. The O\emissiontype{VIII} line is much stronger than the
260: O\emissiontype{VII} line and the Ne\emissiontype{IX} line is so than the
261: Ne\emissiontype{X} line if we taking the energy dependence of the efficiencies into
262: account. These line ratios alone infer that the plasma temperature is in the range of
263: 2.5--4~MK \citep{tucker66} even without spectral model fittings.
264:
265:
266: \subsubsection{Spatially-Resolved Emission}
267: In order to investigate the spatial difference of the plasma properties, we divided the
268: source region into three sub-regions (a, b, and c in figure \ref{fg:f1}a) based on the
269: morphology of the diffuse emission. We constructed the spectra from each region and
270: subtracted the background in the same manner for the entire emission. The merged FI and
271: the BI spectra in each sub-region are shown in figure~\ref{fg:f3}.
272:
273: \begin{figure}
274: \begin{center}
275: \FigureFile(85mm,240mm){f3.eps}
276: \end{center}
277: \caption{Background-subtracted 0.4--1.8~keV band spectra of the diffuse emission in the
278: three regions (a), (b), and (c) in figure \ref{fg:f1} (a). The BI spectrum is shown in red,
279: while the merged FI spectrum is in black in a logarithmic scale. The upper panels show
280: the data in crosses and the best-fit models in solid lines, while the lower panels show
281: residuals to the fit.}\label{fg:f3}
282: \end{figure}
283:
284: We fitted the 0.4--1.8 keV spectra with a thin-thermal plasma model at a collisional
285: equilibrium (the APEC model; \cite{smith01}) convolved with the interstellar absorption
286: \citep{morrison83}. The abundances of the noticeable elements (O, Ne, Mg, and Fe) were
287: free parameters. Those of the other elements were fixed at 0.3 solar, which is
288: canonically used in X-ray spectroscopy in star-forming regions (e.g.,
289: \cite{getman05}). We used the RMFs (version 2006-08-01) of the observation month and
290: generated ARFs using a ray-tracing simulator (\texttt{xissimarfgen} version 2006-08-28;
291: \cite{ishisaki07}) assuming that the emission is uniform across a 15\arcmin\ radius
292: circle centered at the optical axis. In order to compensate for the possible uncertainty
293: in the energy gain calibration, we introduced an additional fitting parameter
294: (offset). The resultant offset values were 1--4 eV, which are within the current
295: calibration limitation.
296:
297: A single temperature model yielded acceptable fits for all spectra. The best-fit
298: hydrogen-equivalent column density ($N_{\rm H}$), plasma temperature ($k_{\rm{B}}T$),
299: metallicity ($Z_{\rm{O}}$, $Z_{\rm{Ne}}$, $Z_{\rm{Mg}}$, and $Z_{\rm{Fe}}$), the average
300: surface brightness ($S_{\rm{X}}$), and the luminosity ($L_{\rm{X}}$) in the 0.5--2.0~keV
301: band are summarized in table~\ref{tb:t1}. The best-fit models are shown in
302: figure~\ref{fg:f3}. We attempted different plasma models with multiple temperatures or\
303: non-equilibrium ionization, but did not obtain improved fits. We therefore consider that
304: a single temperature model at a collisional equilibrium is adequate.
305:
306:
307: \subsection{Spectral Analysis of Discrete Sources}
308: We also constructed spectra of the three discrete sources (1WGA\,J1820.6--1615, M17
309: North, and the OB association in figure~\ref{fg:f1}b). The source signals were accumulated
310: from elliptical regions of 2\farcm0--3\farcm5 axis lengths, while the background signals
311: were from adjacent regions free of sources. The spectra are shown in
312: figure~\ref{fg:f4}. All the spectra are characterized by hard emission full of K$\alpha$
313: emission lines of highly ionized ions, which include Mg\emissiontype{XI},
314: Mg\emissiontype{XII}, Si\emissiontype{XIII}, Si\emissiontype{XIV}, S\emissiontype{XV},
315: S\emissiontype{XVI}, Ar\emissiontype{XVII}, Ar\emissiontype{XVIII},
316: Ca\emissiontype{XIX}, and Fe\emissiontype{XXV}.
317:
318: \begin{figure}
319: \begin{center}
320: \FigureFile(85mm,240mm){f4.eps}
321: \end{center}
322: \caption{Background-subtracted 0.8--10~keV band spectra of the three discrete sources:
323: (a) 1WGA\,J1820.6--1615, (b) M17 North, and (c) the OB association. The symbols follow
324: figure~\ref{fg:f3}. The higher and lower temperature components of the two temperature
325: model are shown respectively with dashed and dotted lines. For simplicity, only the
326: merged FI spectra and the best-fit models are shown.}\label{fg:f4}
327: \end{figure}
328:
329: We can have a crude estimate for spectral models from the intensity of these lines. The
330: spectra of 1WGA~J1820.6--1615 and the OB association show a Mg\emissiontype{XI} line
331: stronger than Mg\emissiontype{XII} as well as Fe\emissiontype{XXV} and a hard continuum
332: up to $\sim$8~keV, requiring at least two thermal components of different
333: temperatures. M17 North shows a Mg\emissiontype{XII} line and no prominent
334: Mg\emissiontype{XI} line, thus a single temperature model would be adequate.
335:
336: We fitted these spectra using the attenuated thin-thermal plasma model similarly for the
337: diffuse emission. The merged FI and the BI spectra were simultaneously fitted for
338: 1WGA\,J1820.6--1615 and the OB association. Only the merged FI spectrum was fitted for
339: the M17 North because its BI spectrum is strongly contaminated by the \atom{Fe}{}{55}
340: calibration source. We first fitted the spectra with a one-temperature model. An
341: additional component with a different temperature was added if the fitting was rejected
342: due to systematic residuals. As the crude estimate, the spectra of 1WGA\,J1820.6--1615
343: and the OB association required a two-temperature model, while that of the M17 North was
344: fitted by a one-temperature model. The best-fit models and parameters are shown in
345: figure~\ref{fg:f4} and table~\ref{tb:t2}, respectively.
346:
347: \begin{table*}
348: \begin{center}
349: \caption{Best-fit APEC parameters for the spatially-resolved diffuse spectra.}\label{tb:t1}
350: \begin{tabular}{llccc}
351: \hline
352: \hline
353: Pars. & Units & \multicolumn{3}{c}{\hrulefill\hspace*{5mm}Sub-regions\hspace*{5mm}\hrulefill} \\
354: & & (a) & (b) & (c) \\
355: \hline
356: $N_{\rm H}$\footnotemark[$*$] & (10$^{21}$~cm$^{-2}$) \dotfill
357: & 4.8 (4.4--5.2) & 4.6 (4.4--5.2) & 4.3 (3.7--4.6) \\
358: $k_{\rm{B}}T$\footnotemark[$*$] & (keV) \dotfill
359: & 0.25 (0.24--0.27) & 0.24 (0.22--0.25) & 0.27 (0.25--0.28) \\
360: $Z_{\rm{O}}$\footnotemark[$*$] & (solar) \dotfill
361: & 0.10 (0.06--0.14) & 0.15 (0.11--0.18) & 0.13 (0.10--0.16) \\
362: $Z_{\rm{Ne}}$\footnotemark[$*$] & (solar) \dotfill
363: & 0.20 (0.17--0.24) & 0.32 (0.25--0.39) & 0.22 (0.19--0.26) \\
364: $Z_{\rm{Mg}}$\footnotemark[$*$] & (solar) \dotfill
365: & 0.10 (0.09--0.13) & 0.12 (0.07--0.16) & 0.12 (0.09--0.16) \\
366: $Z_{\rm{Fe}}$\footnotemark[$*$] & (solar) \dotfill
367: & 0.10 (0.08--0.13) & 0.19 (0.14--0.20) & 0.12 (0.11--0.14) \\
368: $S_{\rm{X}}$\footnotemark[$*\dagger$] & (10$^{-14}$ erg~s$^{-1}$~cm$^{-2}$~arcmin$^{-2})$ \dotfill
369: & 2.16 (2.12--2.21) & 1.34 (1.31--1.38) & 1.55 (1.52--1.58) \\
370: $L_{\rm{X}}$\footnotemark[$\ddagger$] & (10$^{33}$~erg~s$^{-1}$) \dotfill
371: & 1.3 & 1.0 & 1.2\\
372: $\chi^{2}$/d.o.f & \dotfill
373: & 155.5/146 & 121.8/114 & 176.9/129 \\
374: \hline
375: %
376: \multicolumn{5}{@{}l@{}}{\hbox to 0pt{\parbox{170mm}{\footnotesize
377: \par\noindent
378: \footnotemark[$*$] The uncertainties in the parentheses are the 90\% confidence range.
379: \par\noindent
380: \footnotemark[$\dagger$] The average X-ray surface brightness in the 0.5--2.0~keV band.
381: \par\noindent
382: \footnotemark[$\ddagger$] The absorption-corrected X-ray luminosity in the 0.5--2.0~keV band. A distance of 1.6~kpc is assumed.
383: }\hss}}
384: \end{tabular}
385: \end{center}
386: \end{table*}
387:
388: \begin{table*}
389: \begin{center}
390: \caption{Best-fit APEC parameters for the discrete spectra.}\label{tb:t2}
391: \begin{tabular}{llccc}
392: \hline
393: \hline
394: Pars. & Units & 1WGA\,J1820.6--01615 & M17 North & OB association \\
395: \hline
396: $N_{\rm H}$\footnotemark[$*$] & (10$^{22}$~cm$^{-2}$) \dotfill
397: & 1.7 (1.4--2.0) & 2.4 (2.1--2.7) & 1.3 (1.2--1.4) \\
398: $k_{\rm{B}}T_{\rm high}$\footnotemark[$*\dagger$] & (keV) \dotfill
399: & 3.8 (3.1--4.7) &2.7 (2.4--3.2) & 4.0 (3.9--4.1) \\
400: $k_{\rm{B}}T_{\rm low}$\footnotemark[$*\dagger$] & (keV) \dotfill
401: & 0.56 (0.47--0.65) & ... &0.59 (0.56--0.62) \\
402: $Z_{\rm{Ne}}$\footnotemark[$*$] & (solar) \dotfill
403: & 0.3 &0.3 & 0.51 (0.34--0.71) \\
404: $Z_{\rm{Mg}}$\footnotemark[$*$] & (solar) \dotfill
405: & 0.49 (0.24--0.87) & 0.3 & 0.51 (0.38--0.65) \\
406: $Z_{\rm{Si}}$\footnotemark[$*$] & (solar) \dotfill
407: & 0.30 (0.16--0.58) & 0.3 & 0.53 (0.39--0.71) \\
408: $Z_{\rm{S}} $\footnotemark[$*$] & (solar) \dotfill
409: & 1.07 (0.58--1.72) & 0.3 & 1.17 (0.98--1.38) \\
410: $Z_{\rm{Ar}}$\footnotemark[$*$] & (solar) \dotfill
411: & 0.3 & 1.67 (0.72--3.00) & 1.44 (1.00--1.90) \\
412: $Z_{\rm{Ca}}$\footnotemark[$*$] & (solar) \dotfill
413: & 0.3 & 0.3 & 0.46 (0.01--0.90) \\
414: $Z_{\rm{Fe}}$\footnotemark[$*$] & (solar) \dotfill
415: & 0.33 (0.19--0.48) & 0.36 (0.22--0.50) & 0.28 (0.25--0.31) \\
416: $F_{\rm{X}}$\footnotemark[$*\ddagger$] & (10$^{-13}$ erg~s$^{-1}$~cm$^{-2}$) \dotfill
417: & 3.9 (3.8--4.0) & 7.4 (7.1--7.7) & 58.3 (57.8--58.8) \\
418: $L_{\rm{X}}$\footnotemark[$\S$]& (10$^{32}$~erg~s$^{-1}$) \dotfill
419: & 3.2 & 4.9 & 30.0 \\
420: $\chi^2$/d.o.f. & \dotfill
421: & 100.2/134 & 66.4/76 & 473.3/405 \\
422: \hline
423: %
424: \multicolumn{5}{@{}l@{}}{\hbox to 0pt{\parbox{150mm}{\footnotesize
425: \par\noindent
426: \footnotemark[$*$] The uncertainties in the parentheses are the 90\% confidence
427: range. Fixed values are shown without ranges.
428: \par\noindent
429: \footnotemark[$\dagger$] The plasma temperatures for the higher and lower temperature
430: components. Only the higher temperature value is given for M17 North, which is fitted
431: by a single temperature model.
432: \par\noindent
433: \footnotemark[$\ddagger$] The X-ray flux in the 1.0--8.0~keV band.
434: \par\noindent
435: \footnotemark[$\S$] The absorption-corrected X-ray luminosity in the 1.0--8.0~keV band. A distance of 1.6~kpc is assumed.
436: }\hss}}
437: \end{tabular}
438: \end{center}
439: \end{table*}
440:
441:
442: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
443: \section{Discussion}
444: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
445: \subsection{Contamination to the Extended Emission}\label{sect:s4-1}
446: The spectra of the diffuse emission in figures~\ref{fg:f2} and \ref{fg:f3} are
447: contaminated by other sources of emission. We evaluate the levels of contamination by
448: unresolved point sources, the Galactic Ridge X-ray emission (GRXE), the cosmic X-ray
449: background (CXB), and the local hot bubble (LHB), and argue that their contributions to
450: the background-subtracted spectra are negligible. About 70\% of the XIS field is covered
451: in the Chandra observation (figure~\ref{fg:f1}b), which has a much better spatial
452: resolution and sensitivity for faint point sources \citep{townsley03,broos07}. Among the
453: three sub-regions, the region (a) has a complete coverage by Chandra. We therefore use
454: this sub-region as a representative to evaluation the levels of various contaminations.
455:
456: First, we examine the contribution of unresolved point sources. We extracted 19 point
457: sources from the ACIS data in the sub-region (a), constructed the composite spectrum,
458: and fitted it with a thermal plasma model. The spectrum of each source is too poor to
459: fit individually, so we assume that all sources have the same spectral shape that best
460: describes the composite spectrum. Using their positions, flux, and the assumed spectral
461: shape, we generated their XIS events using a ray-tracing simulator (\texttt{xissim};
462: \cite{ishisaki07}). In figure~\ref{fg:f5}, we compare the integral of the simulated
463: spectra of unresolved point sources (PS1) to the observed diffuse spectrum (pluses). In
464: the displayed observed spectrum, we subtracted the NXB spectrum but not the background
465: spectrum in the neighboring region. The point source contribution accounts for $\sim$8\%
466: of the emission in the 0.4--1.8~keV band.
467:
468: \begin{figure}
469: \begin{center}
470: \FigureFile(85mm,60mm){f5.eps}
471: \end{center}
472: \caption{Comparison of the observed diffuse spectrum to the simulated spectra of
473: contaminating sources. The NXB signal is subtracted from the observed spectrum. PS1
474: and PS2 are for the contribution of unresolved point sources with different spectral
475: models, GRXE for the Galactic Ridge X-ray emission, CXB for the cosmic X-ray
476: background, and LHB for the local hot bubble. None of them play a significant
477: contribution. Moreover, most of them are removed by subtracting a background
478: spectrum from the neighboring region.}\label{fg:f5}
479: \end{figure}
480:
481: To have an estimate of the maximum contamination to the lines, we repeated the same
482: procedure using the model with the maximum allowable abundance values in the best-fit
483: model for simulating unresolved point source events. The resultant integrated spectrum
484: (PS2) is also shown in figure~\ref{fg:f5}. At the Ne\emissiontype{IX} line at 0.92~keV,
485: the contribution by the point sources accounts only for $<$10\% of the observed
486: emission.
487:
488: Second, we derive the GRXE contribution, which is ubiquitous along the Galactic Plane at
489: Galactic longitudes ($l$) of $|l| \lesssim$~45\arcdeg\ \citep{kaneda97,sugizaki01}. We
490: refer to a Chandra result \citep{ebisawa05} for the spectral shape in the soft band and
491: to a Rossi X-ray Timing Explorer result \citep{revnivtsev06} for the surface brightness
492: at the position of M17 ($l \sim$~15\arcdeg). The GRXE contribution to the observed
493: emission is estimated to be $\sim$3\% and is shown in figure~\ref{fg:f5}.
494:
495: Third, for the CXB contribution, we consulted the Suzaku XIS observation of the North
496: Ecliptic Pole \citep{fujimoto07} both for the spectral shape and the surface
497: brightness. The CXB was observed and fitted by a power-law model. We convolved the model
498: with the XIS responses and found that the contribution is $\sim$2\%.
499:
500: Finally, we constrain the contribution by the LHB emission. We estimate its surface
501: brightness to be $\sim$4$\times$10$^4$~counts~s$^{-1}$~arcmin$^{-2}$ in the PSPC R1 and
502: R2 bands \citep{snowden98}. We assume that the spectrum is a thin-thermal plasma (the
503: Raymond-Smith model; \cite{raymond77}) with a temperature of 0.1~keV. With the derived
504: emission measure of $\sim$2.8$\times10^{-3}$~cm$^{-6}$~pc, the LHB contributes $\sim$2\%
505: of the observed emission.
506:
507: None of the above components play a significant role in the observed diffuse
508: spectrum. Moreover, we removed most of them by subtracting a background spectrum from a
509: neighboring region. The background events show no indication of time variability
510: caused by solar flares \citep{fujimoto07}. We conclude that the background-subtracted
511: XIS spectrum represents the spectrum of the uncontaminated diffuse emission quite well.
512:
513: The background spectrum accounts for $\sim$35\% of the source spectrum of the sub-region
514: (a). Besides the NXB contributing $\sim$7\% of the source flux, the sum of unresolved
515: point sources, GRXE, CXB, and LHB contributes $\sim$15\%. Therefore, emission with a
516: flux of $\sim$13\% of the source flux is additionally included in the background
517: spectrum. The contribution from the bright sources outside of the extraction region is
518: negligible. We attribute the remaining emission to the diffuse emission in the
519: background region (figure~\ref{fg:f1}a). If this is the case, the flux estimate of the
520: diffuse emission (table~\ref{tb:t1}) is underestimated by $\sim$13\%.
521:
522:
523: \subsection{Comparison with Previous Studies}\label{sect:s4-2}
524: We compare our results with the previous works using ROSAT \citep{dunne03} and Chandra
525: \citep{townsley03,broos07}. We derived that the diffuse spectrum is explained by an
526: absorbed single temperature thin-thermal plasma model of $k_{\rm{B}}T \sim$~0.25~keV and
527: $N_{\rm H} \sim$~4.5$\times$10$^{21}$~cm$^{-2}$. The total luminosity (0.5--2.0~keV) in
528: the combined (a)$+$(b)$+$(c) region is
529: $\sim$3.5$\times$10$^{33}$~erg~s$^{-1}$ (table~\ref{tb:t1}).
530:
531: The ROSAT study \citep{dunne03} shows that the total luminosity of the diffuse emission
532: is $\sim$2.5$\times$10$^{33}$~erg~s$^{-1}$. The smaller estimate than our result is more
533: noticeable if we consider that the ROSAT value was derived from a larger area and in a
534: wider energy range (0.1--2.4~keV). This stems from an underestimate of the
535:
536: extinction. Without a sufficient spectral resolution to resolve lines, two different
537: models were not disentangled in the ROSAT PSPC spectra; one is a low plasma temperature
538: with a large extinction ($\sim 0.2$~keV and $\sim 10^{22}$~cm$^{-2}$) and the other is a
539: high plasma temperature with a small extinction ($\sim$0.7~keV and $\sim
540: 10^{20}-10^{21}$~cm$^{-2}$). \citet{dunne03} derived the luminosity based on the latter,
541: but our spectroscopy shows that the former should be in the case.
542:
543: The best-fit XIS values of $L_{\rm{X}}$ and $N_{\rm H}$ are consistent with those
544: presented in the Chandra study \citep{townsley03}, in which
545: $L_{\rm{X}}=$~3.4$\times$10$^{33}$~erg~s$^{-1}$ (0.5--2.0~keV), and $N_{\rm
546: H}=$~(4~$\pm$~1)$\times$10$^{21}$~cm$^{-2}$. However, the plasma temperatures are
547: different between the two studies. In Chandra, the primary component of
548: $k_{\rm{B}}T=$~0.6~$\pm$~0.1~keV and the secondary component of $\sim$0.13~keV were
549: claimed. In Suzaku, however, we confirmed that a single temperature component of
550: $\sim$0.25~keV is adequate from the diagnosis of resolved oxygen and neon lines.
551:
552: From the discrete sources, we detected K$\alpha$ emission lines from highly ionized ions
553: (figure~\ref{fg:f4}). The spectrum of the OB association is comprised of hundreds of
554: point sources, but the emission from an O4--O5 binary dominates the
555: spectrum. \citet{broos07} claimed that both components of the binary (sources 543 and
556: 536) have plasma temperatures exceeding 10~keV. However, the strong Fe\emissiontype{XXV}
557: K$\alpha$ line at 6.7~keV and the weak Fe\emissiontype{XXVI} K$\alpha$ line at 7.0~keV
558: in the XIS spectrum (figure~\ref{fg:f4}c) do not support such high temperatures.
559:
560:
561: \subsection{Spatial Difference of the Plasma Properties}\label{sect:s4-3}
562: A high signal-to-noise ratio spectrum by XIS enabled us to conduct spatially-resolved
563: spectroscopy of the diffuse emission. The plasma temperature and the chemical
564: composition are uniform, except possibly for a larger metallicity in the sub-region
565: (b). The Chandra study \citep{townsley03} also show no evidence for spatial variation of
566: plasma temperature.
567:
568: The observed uniformity indicates that the entire plasma is at a thermal equilibrium in
569: the observed spatial scale of $\sim$5~pc, unless the plasma is patchy at equilibria
570: locally by magnetic confinement. The global equilibrium is reasonable considering the
571: fact that the plasma sound crossing time ($\sim$2$\times$10$^{4}$~yr) is much smaller
572: than the time scale of the system ($\sim$10$^{6}$~yr), thus the constant pressure is
573: achieved \citep{weaver77}. Here, we used the plasma volume and the electron density
574: as $\sim$30~pc$^{3}$ and $\sim$1~cm$^{-3}$, respectively, by assuming that the plasma
575: distribution has a conical shape with its apex at the OB association and with a filling
576: factor of 1. Given the uniformity of the plasma temperature and pressure, we speculate
577: that the density is also spatially uniform.
578:
579: The observed surface brightness, however, is different among the three sub-regions
580: (figure~\ref{fg:f1}a, table~\ref{tb:t1}). It is $\sim$1.6 times more intense in the
581: sub-region (a) than (b). This is not attributable entirely to the different extinction,
582: as $N_{\rm H}$ is larger in the sub-region (a) than in (b). Because the plasma has a
583: uniform temperature and density, we speculate that the difference of the surface
584: brightness is likely due to the different line-of-sight depths or different filling
585: factors.
586:
587: With the derived plasma volume and the density, the total mass of the plasma is
588: $\sim$1~$M_{\odot}$. This is comparable to the integrated ejecta mass by stellar winds
589: at a mass loss rate of $\sim$10$^{-6}$~$M_{\odot}$ for $\sim$10$^{6}$~yr and agrees with
590: the estimates by the previous works \citep{dunne03,townsley03}. We also speculate that
591: the swept-up and evaporated ISM does not make a significant contribution to the plasma
592: mass.
593:
594:
595: \subsection{Chemical Composition}\label{sect:s4-4}
596: The chemical composition of the diffuse emission is revealed for the first time in this
597: study. The metallicity of the diffuse emission is 0.1--0.3 solar (table~\ref{tb:t1}),
598: which is significantly lower than those derived for the discrete sources (0.3--1.5
599: solar). The metal abundances relative to each other in the diffuse emission are
600: consistent with solar among O, Mg, and Fe \citep{anders89}, but Ne is enhanced in all
601: three sub-regions by a factor of $\sim$2 (table~\ref{tb:t1}). This is also evident in
602: the simultaneous spectral fits of the three sub-regions, in which we tied the abundance
603: values of these elements. The resultant values are $Z_{\rm{O}}=$~0.12 (0.11--0.13),
604: $Z_{\rm{Ne}}=$~0.22 (0.21--0.25), $Z_{\rm{Mg}}=$~0.10 (0.08--0.12), and
605: $Z_{\rm{Fe}}=$~0.12 (0.11--0.13) solar.
606:
607: Such Ne enhancement from other metals is widely seen in coronally active stars
608: \citep{brinkman01,kastner02,audard03,imanishi03,stelzer04,maggio07} for unknown
609: reasons. We consider that the Ne enhancement is an intrinsic feature of the diffuse
610: plasma, and is not influenced by the contamination of Ne-enhanced point sources spectra
611: (\S~\ref{sect:s4-1}). One explanation for the anomaly is that the poorly-constrained
612: solar Ne abundance is underestimated by a factor of a few \citep{drake05,liefke06}. This
613: would account for the observed Ne enhancement in the M17 diffuse plasma as well.
614:
615: The lack of a clear spatial variation of the chemical composition comprises a sharp
616: contrast to the diffuse emission in the Carina Nebula \citep{hamaguchi07}, where the
617: different abundance patterns across a similar spatial scale suggest the supernova origin
618: for the emission. In the diffuse emission in M17, the O and Fe ratio is consistent with
619: the solar abundance, which is another line of evidence against the supernova
620: interpretation. If a supernova has occurred in M17, it should have been caused by a
621: star earlier than the earliest (O4--O5) star in the OB association. Such a massive
622: source causes a core-collapse--type supernova. It would have yielded a measurably larger
623: ratio of O against Fe than the solar value by a factor of a few
624: \citep{tsujimoto95,nomoto97}.
625:
626:
627: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
628: \section{Summary}
629: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
630: We conducted a spectroscopic study of the soft diffuse X-ray emission in M17 using the
631: XIS onboard Suzaku. High signal-to-noise ratio spectra of the diffuse emission were
632: obtained in a spatially-resolved manner. Half a dozen of emission lines were resolved
633: clearly for the first time, which include K$\alpha$ lines from highly ionized O, Ne, and
634: Mg and L lines from Fe. These lines are clear evidence for the thermal origin of the
635: diffuse emission.
636:
637: Based on the Chandra data of an overlapping field and on the previous works in the
638: literature, we confirmed that the background-subtracted diffuse spectrum by XIS is
639: barely contaminated by unresolved point sources, GRXE, CXB, and LHB emission.
640:
641: We showed that the diffuse spectra are explained by a single temperature plasma model of
642: $k_{\rm{B}}T \sim$~0.25~keV, $L_{\rm{X}} \sim$~3.5$\times$10$^{33}$~erg~s$^{-1}$
643: (0.5--2.0~keV), and $N_{\rm H} \sim$4.5$\times$10$^{21}$~cm$^{-2}$. The temperature and
644: the chemical composition of the diffuse plasma is spatially uniform, indicating that the
645: plasma is at a thermal equilibrium. The apparent difference in the surface brightness is
646: probably due to the difference in the line-of-sight depth or in the filling factors, and
647: not in the plasma density.
648:
649: The abundance is obtained individually for the detected elements, which are consistent
650: with 0.1--0.3~solar values. The enhancement of Ne against other metals by a factor of
651: $\sim$2 is seen in the diffuse plasma, which may be explained by a upward revision of
652: the solar Ne abundance.
653:
654: The lack of spatial variations in the chemical composition comprises a sharp contrast to
655: the diffuse emission in the Carina Nebula, where different abundance patterns across a
656: similar spatial scale suggest the supernova origin for the emission. Together with the O
657: to Fe ratio consistent with the solar value, this gives evidence against the
658: interpretation that a supernova is the cause of the diffuse emission observed in M17.
659:
660:
661: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
662: % Acknowledgments
663: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
664: \bigskip
665:
666: The authors thank Takashi Hosokawa for useful discussion. Y.\,H. and M.\,T. acknowledge
667: financial support from the Japan Society for the Promotion of Science. The work is
668: supported by the Grants-in-Aid for the 21st century center of excellence program
669: ``Center for Diversity and Universality in Physics'' and for the program number 18204015
670: from the Ministry of Education, Culture, Sports, Science and Technology of Japan. The
671: research made use of data obtained from the Data ARchive and Transmission System
672: (DARTS), provided by the PLAIN center, ISAS/JAXA.
673:
674:
675: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
676: % Bibliography
677: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
678: \begin{thebibliography}{}
679: \bibitem[Anders, Grevesse(1989)]{anders89} Anders, E., \& Grevesse, N. 1989,
680: Geochim. Cosmochim. Acta, 53, 197
681: \bibitem[Audard et al.(2003)]{audard03} Audard, M., G{\"u}del, M., Sres, A., Raassen,
682: A.~J.~J., \& Mewe, R.\ 2003, \aap, 398, 1137
683: \bibitem[Brinkman et al.(2001)]{brinkman01} Brinkman, A.~C., et al.\ 2001, \aap, 365,
684: L324
685: \bibitem[Brogan \& Troland(2001)]{brogan01} Brogan, C.~L., \& Troland, T.~H.\ 2001,
686: \apj, 560, 821
687: \bibitem[Broos et al.(2007)]{broos07} Broos, P.~S., Feigelson, E.~D., Townsley, L.~K.,
688: Getman, K.~V., Wang, J., Garmire, G.~P., Jiang, Z., \& Tsuboi, Y.\
689: 2007, \apjs, 169, 353
690: \bibitem[Cappellari et al.(2003)]{cappellari03} Cappellari, M., \& Copin, Y. 2003,
691: \mnras, 342, 345
692: \bibitem[Chrysostomou et al.(1992)]{chrysostomou02} Chrysostomou, A., Brand,
693: P.~W.~J.~L., Burton, M.~G., \& Moorhouse, A.\ 1992, \mnras, 256, 528
694: \bibitem[Chu et al.(1995)]{chu95} Chu, Y.-H., Chang, H.-W., Su, Y.-L., \& Mac Low,
695: M.-M.\ 1995, \apj, 450, 157
696: \bibitem[Diehl et al.(2006)]{Diehl06} Diehl, S., \& Statler, T. S. 2006, \mnras, 368,
697: 497
698: \bibitem[Drake \& Testa(2005)]{drake05} Drake, J.~J., \& Testa, P.\ 2005, \nat, 436,
699: 525
700: \bibitem[Dunne et al.(2003)]{dunne03} Dunne, B.~C., Chu, Y.-H., Chen, C.-H.~R., Lowry,
701: J.~D., Townsley, L., Gruendl, R.~A., Guerrero, M.~A., \& Rosado, M.\
702: 2003, \apj, 590, 306
703: \bibitem[Ebisawa et al.(2005)]{ebisawa05} Ebisawa, K., et al.\ 2005, \apj, 635, 214
704: \bibitem[Felli et al.(1984)]{felli84} Felli, M., Massi, M., \& Churchwell, E.\ 1984,
705: \aap, 136, 53
706: \bibitem[Fujimoto et al.(2007)]{fujimoto07} Fujimoto, R., et al.\ 2007, \pasj, 59, S133
707: \bibitem[Garmire et al.(2003)]{garmire03} Garmire, G.~P., Bautz, M.~W., Ford, P.~G.,
708: Nousek, J.~A., \& Ricker, G.~R., Jr.\ 2003, \procspie, 4851, 28
709: \bibitem[Getman et al.(2005)]{getman05} Getman, K.~V., et al.\ 2005, \apjs, 160, 319
710: \bibitem[Hamaguchi et al.(2007)]{hamaguchi07} Hamaguchi, K., et al.\ 2007,
711: \pasj, 59, S151
712: \bibitem[Hanson et al.(1997)]{hanson97} Hanson, M.~M., Howarth, I.~D., \& Conti, P.~S.\
713: 1997, \apj, 489, 698
714: \bibitem[Henning et al.(1998)]{henning98} Henning, T., Klein, R., Launhardt, R., Lemke,
715: D., \& Pfau, W.\ 1998, \aap, 332, 1035
716: \bibitem[Imanishi et al.(2003)]{imanishi03} Imanishi, K., Nakajima, H., Tsujimoto, M.,
717: Koyama, K., \& Tsuboi, Y.\ 2003, \pasj, 55, 653
718: \bibitem[Ishisaki et al.(2007)]{ishisaki07} Ishisaki, Y., et al.\ 2007, \pasj, 59, S113
719: \bibitem[Jiang et al.(2002)]{jiang02} Jiang, Z., et al.\ 2002, \apj, 577, 245
720: \bibitem[Kaneda et al.(1997)]{kaneda97} Kaneda, H., Makishima, K., Yamauchi, S.,
721: Koyama, K., Matsuzaki, K., \& Yamasaki, N.~Y.\ 1997, \apj, 491, 638
722: \bibitem[Kastner et al.(2002)]{kastner02} Kastner, J.~H., Huenemoerder, D.~P., Schulz,
723: N.~S., Canizares, C.~R., \& Weintraub, D.~A.\ 2002, \apj, 567, 434
724: \bibitem[Kokubun et al.(2007)]{kokubun07} Kokubun, M., et al.\ 2007, \pasj, 59, S53
725: \bibitem[Koyama et al.(2007)]{koyama07} Koyama, K., et al.\ 2007, \pasj, 59, S23
726: \bibitem[Lada et al.(1991)]{lada91} Lada, C.~J., Depoy, D.~L., Merrill, K.~M., \&
727: Gatley, I.\ 1991, \apj, 374, 533
728: \bibitem[Leitherer et al.(1992)]{leitherer92} Leitherer, C., Robert, C., \& Drissen,
729: L.\ 1992, \apj, 401, 596
730: \bibitem[Liefke \& Schmitt(2006)]{liefke06} Liefke, C., \& Schmitt, J.~H.~M.~M.\ 2006,
731: \aap, 458, L1
732: \bibitem[Maggio et al.(2007)]{maggio07} Maggio, A., Flaccomio, E., Favata, F., Micela,
733: G., Sciortino, S., Feigelson, E.~D., \& Getman, K.~V.\ 2007, \apj, 660,
734: 1462
735: \bibitem[Mitsuda et al.(2007)]{mitsuda07} Mitsuda, K., et al.\ 2007, \pasj, 59, S1
736: \bibitem[Morrison \& McCammon(1983)]{morrison83} Morrison, R., \& McCammon, D.\ 1983,
737: \apj, 270, 119
738: \bibitem[Nielbock et al.(2001)]{nielbock01} Nielbock, M., Chini, R., J{\"u}tte, M., \&
739: Manthey, E.\ 2001, \aap, 377, 273
740: \bibitem[Nomoto et al.(1997)]{nomoto97} Nomoto, K., Hashimoto, M., Tsujimoto, T.,
741: Thielemann, F.-K., Kishimoto, N., Kubo, Y., \& Nakasato, N.\ 1997,
742: Nuclear Physics A, 616, 79
743: \bibitem[Ogura \& Ishida(1976)]{ogura76} Ogura, K., \& Ishida, K. 1976, \pasj, 28, 35
744: \bibitem[Pfeffermann et al.(1987)]{pfeffermann87} Pfeffermann, E., et al.\ 1987,
745: \procspie, 733, 519
746: \bibitem[Povich et al.(2007)]{povich07} Povich, M.~S., et al.\ 2007, \apj, 660, 346
747: \bibitem[Raymond \& Smith(1977)]{raymond77} Raymond, J.~C., \& Smith, B.~W.\ 1977,
748: \apjs, 35, 419
749: \bibitem[Revnivtsev et al.(2006)]{revnivtsev06} Revnivtsev, M., Molkov, S., \& Sazonov,
750: S.\ 2006, \mnras, 373, L11
751: \bibitem[Serlemitsos et al.(2007)]{serlemitsos07} Serlemitsos, P.~J., et al.\ 2007,
752: \pasj, 59, S9
753: \bibitem[Smith et al.(2001)]{smith01} Smith, R.~K., Brickhouse, N.~S., Liedahl, D.~A.,
754: \& Raymond, J.~C.\ 2001, \apjl, 556, L91
755: \bibitem[Snowden et al.(1998)]{snowden98} Snowden, S.~L., Egger, R., Finkbeiner, D.~P.,
756: Freyberg, M.~J., \& Plucinsky, P.~P.\ 1998, \apj, 493, 715
757: \bibitem[Stelzer \& Schmitt(2004)]{stelzer04} Stelzer, B., \& Schmitt, J.~H.~M.~M.\
758: 2004, \aap, 418, 687
759: \bibitem[Sugizaki et al.(2001)]{sugizaki01} Sugizaki, M., Mitsuda, K., Kaneda, H.,
760: Matsuzaki, K., Yamauchi, S., \& Koyama, K.\ 2001, \apjs, 134, 77
761: \bibitem[Takahashi et al.(2007)]{takahashi07} Takahashi, T., et al.\ 2007, \pasj, 59,
762: S35
763: \bibitem[Townsley et al.(2003)]{townsley03} Townsley, L.~K., Feigelson, E.~D.,
764: Montmerle, T., Broos, P.~S., Chu, Y.-H., \& Garmire, G.~P.\ 2003, \apj,
765: 593, 874
766: \bibitem[Tr\"{u}mper(1982)]{trumper82} Tr\"{u}mper, J.\ 1982, Advances in Space Research,
767: 2, 241
768: \bibitem[Tsujimoto et al.(1995)]{tsujimoto95} Tsujimoto, T., Nomoto, K., Yoshii,
769: Y., Hashimoto, M., Yanagida, S., \& Thielemann, F.-K.\ 1995, \mnras,
770: 277, 945
771: \bibitem[Tsujimoto et al.(2007)]{tsujimoto07} Tsujimoto, M., Hyodo, Y., \& Koyama, K.\
772: 2007, \pasj, 59, S229
773: \bibitem[Tucker \& Gould(1966)]{tucker66} Tucker, W.~H., \& Gould, R.~J.\ 1966, \apj,
774: 144, 244
775: \bibitem[Weaver et al.(1977)]{weaver77} Weaver, R., McCray, R., Castor, J., Shapiro,
776: P., \& Moore, R.\ 1977, \apj, 218, 377
777: \bibitem[Weisskopf et al.(2002)]{weisskopf02} Weisskopf, M.~C., Brinkman, B.,
778: Canizares, C., Garmire, G., Murray, S., \& van Speybroeck, L.~P.\ 2002,
779: \pasp, 114, 1
780: \bibitem[White et al.(1994)]{white94} White, N.~E., Giommi, P., \& Angelini, L.\ 1994,
781: \iaucirc, 6100, 1
782: \bibitem[Wilson et al.(1979)]{wilson79} Wilson, T.~L., Fazio, G.~G., Jaffe, D.,
783: Kleinmann, D., Wright, E.~L., \& Low, F.~J.\ 1979, \aap, 76, 86
784: \end{thebibliography}
785: \end{document}
786: