astro-ph0608163/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: 
3: %% manuscript produces a one-column, double-spaced document:
4: 
5: %%\documentclass[manuscript]{aastex}
6: 
7: %% preprint2 produces a double-column, single-spaced document:
8: 
9: %%\documentclass[preprint2]{aastex}
10: 
11: %% If you want to create your own macros, you can do so
12: %% using \newcommand. Your macros should appear before
13: %% the \begin{document} command.
14: %%
15: %% If you are submitting to a journal that translates manuscripts
16: %% into SGML, you need to follow certain guidelines when preparing
17: %% your macros. See the AASTeX v5.x Author Guide
18: %% for information.
19: 
20: %\usepackage{natbib}
21: %\usepackage{lscape}
22: 
23: \newcommand{\vdag}{(v)^\dagger}
24: \newcommand{\myemail}{ezoe@amalthea.phys.s.u-tokyo.ac.jp}
25: 
26: %
27: % instruments
28: %
29: %\def\Astro-E{{\it Astro-E} }
30: %\def\AXAF{{\it AXAF} }
31: %\def\ASCA{{\it ASCA} }
32: %\def\Chandra{{\it Chandra} }
33: %\def\Einstein{{\it Einstein} }
34: %\def\EXOSAT{{\it EXOSAT} }
35: %\def\Ginga{{\it Ginga} }
36: %\def\HEAO-1{{\it HEAO}-1 }
37: %\def\Newton{{\it Newton} }
38: %\def\RXTE{{\it RXTE} }
39: %\def\ROSAT{{\it ROSAT} }
40: %\def\Tenma{{\it Tenma} }
41: %\def\XMM{{\it XMM} }
42: %
43: 
44: %
45: % Misc
46: %
47: \def\degree{$^{\circ}$ }
48: \def\ergs{ergs s$^{\rm -1}$ }
49: \def\flux{ergs s$^{\rm -1}$ cm s$^{\rm -2}$ }
50: 
51: %
52: % astrophysical constants
53: %
54: \def\msolar{\mbox{M$_{\odot}$} }
55: \def\lsolar{\mbox{L$_{\odot}$} }
56: \def\rsolar{\mbox{R$_{\odot}$} }
57: 
58: %
59: % mathematical expressions
60: %
61: \def\ltsima{$\; \buildrel < \over \sim \;$}
62: \def\simlt{\lower.5ex\hbox{\ltsima}}            % < over ~
63: \def\gtsima{$\; \buildrel > \over \sim \;$}
64: \def\simgt{\lower.5ex\hbox{\gtsima}}            % > over ~
65: 
66: %
67: % other macros
68: %
69: \def\beq{\begin{equation}}
70: \def\eeq{\end{equation}}
71: \def\beqary{\begin{eqnarray}}
72: \def\eeqary{\end{eqnarray}}
73: %
74: \newcommand{\eqlabel}[1]{\label{eq:#1}}
75: \newcommand{\figlabel}[1]{\label{fig:#1}}
76: \newcommand{\tablabel}[1]{\label{tab:#1}}
77: \newcommand{\chaplabel}[1]{\label{chap:#1}}
78: \newcommand{\seclabel}[1]{\label{sec:#1}}
79: %
80: \newcommand{\eqref}[1]{(\ref{eq:#1})}
81: \newcommand{\figref}[1]{Figure \ref{fig:#1}}
82: \newcommand{\tabref}[1]{Table \ref{tab:#1}}
83: \newcommand{\chapref}[1]{Chapter \ref{sec:#1}}
84: \newcommand{\secref}[1]{Section \ref{sec:#1}}
85: 
86: %% You can insert a short comment on the title page using the command below.
87: 
88: %\slugcomment{Not to appear in Nonlearned J., 45.}
89: 
90: %% If you wish, you may supply running head information, although
91: %% this information may be modified by the editorial offices.
92: %% The left head contains a list of authors,
93: %% usually a maximum of three (otherwise use et al.).  The right
94: %% head is a modified title of up to roughly 44 characters.  Running heads
95: %% will not print in the manuscript style.
96: 
97: %%\shorttitle{Collapsed Cores in Globular Clusters}
98: %%\shortauthors{Djorgovski et al.}
99: 
100: %% This is the end of the preamble.  Indicate the beginning of the
101: %% paper itself with \begin{document}.
102: 
103: \begin{document}
104: 
105: %% LaTeX will automatically break titles if they run longer than
106: %% one line. However, you may use \\ to force a line break if
107: %% you desire.
108: 
109: \title{ Discovery of Diffuse X-ray Emission in One of
110:         the Nearest Massive Star-Forming Regions NGC 2024}
111: 
112: %% Use \author, \affil, and the \and command to format
113: %% author and affiliation information.
114: %% Note that \email has replaced the old \authoremail command
115: %% from AASTeX v4.0. You can use \email to mark an email address
116: %% anywhere in the paper, not just in the front matter.
117: %% As in the title, you can use \\ to force line breaks.
118: 
119: \author{Y. Ezoe\altaffilmark{1,2},
120:         M. Kokubun\altaffilmark{1},
121:         K. Makishima\altaffilmark{1,3},
122:         Y. Sekimoto\altaffilmark{4},
123:         K. Matsuzaki\altaffilmark{2}}
124: 
125: %% Notice that each of these authors has alternate affiliations, which
126: %% are identified by the \altaffilmark after each name.  Specify alternate
127: %% affiliation information with \altaffiltext, with one command per each
128: %% affiliation.
129: 
130: \altaffiltext{1}{Department of Physics, University of Tokyo,
131:                  7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan}
132: \altaffiltext{2}{The Institute of Space and Astronautical Science,
133:                  3-1-1 Yoshinodai, Sagamihara, Kanagawa 229-8510, Japan}
134: \altaffiltext{3}{RIKEN (The Institute of Physical and Chemical Research),
135:                  2-1, Hirosawa, Wako, Saitama 351-0198, Japan}
136: \altaffiltext{4}{National Astronomical Observatory of Japan,
137:                  2-21-1 Osawa, Mitaka, Tokyo, 181-8588, Japan}
138: 
139: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
140: %% style, abstract will output a Received/Accepted line after the
141: %% title and affiliation information. No date will appear since the author
142: %% does not have this information. The dates will be filled in by the
143: %% editorial office after submission.
144: 
145: \begin{abstract}
146: 
147: A deep 75 ks {\it Chandra} ACIS--I data of NGC 2024 was analyzed, aiming 
148: at a search for diffuse X-ray emission in this one of the most nearby 
149: (415 pc) massive star-forming regions. After removing point sources, 
150: an extended emission was detected in the central circular region with a 
151: radius of 0.5 pc. 
152: It is spatially associated with the young massive stellar cluster. 
153: Its X-ray spectrum exhibits a very hard continuum ($kT>8$ keV)
154: and a sign 
155: of He-like Fe K$_\alpha$ line with the 0.5--7 keV absorption corrected 
156: luminosity of 2$\times10^{31}$ \ergs.
157: Undetected faint point sources, estimated from the luminosity function of
158: the detected sources, contribute less than 10\% to this emission.
159: Hence the emission is truly diffuse in nature. Because of the proximity of 
160: NGC 2024 and the long exposure, this discovery is one of the most strong 
161: supports for the existence of the diffuse X-ray emission in massive 
162: star-forming regions.
163: 
164: \end{abstract}
165: 
166: %% Keywords should appear after the \end{abstract} command. The uncommented
167: %% example has been keyed in ApJ style. See the instructions to authors
168: %% for the journal to which you are submitting your paper to determine
169: %% what keyword punctuation is appropriate.
170: 
171: \keywords{HII regions --- ISM: individual (NGC 2024) --- stars: formation ---stars: early-type --- stars: winds, outflows}
172: 
173: %% From the front matter, we move on to the body of the paper.
174: %% In the first two sections, notice the use of the natbib \citep
175: %% and \citet commands to identify citations.  The citations are
176: %% tied to the reference list via symbolic KEYs. The KEY corresponds
177: %% to the KEY in the \bibitem in the reference list below. We have
178: %% chosen the first three characters of the first author's name plus
179: %% the last two numeral of the year of publication as our KEY for
180: %% each reference.
181: 
182: \section{Introduction}
183: \label{sec:intro}
184: 
185: Increasing evidences of diffuse X-ray emission 
186: has been found in massive star-forming regions (MSFRs) with 
187: the {\it Chandra X-ray Observatory}; 
188: the Rosette Nebula (at a distance of $D=1.4$ kpc; \citealt{Townsley2003}), 
189: M17 ($D=1.6$ kpc; \citealt{Townsley2003}), 
190: RCW 38 ($D=1.6$ kpc; \citealt{Wolk2002}), 
191: NGC 6334 ($D=1.7$ kpc; \citealt{Ezoe2006}),
192: RCW 49 ($D=$1.9-7.9 kpc; \citealt{Townsley2004}),
193: W51A ($D=$5.5-7.5 kpc; \citealt{Townsley2004}),
194: NGC 3603 ($D=7$ kpc; \citealt{Moffat2002}),
195: the Arches Cluster ($D=8.5$ kpc; \citealt{Yusef-Zadeh2002}), 
196: and the Quintuplet Cluster ($D=8.5$ kpc; ; \citealt{Wang2002}).
197: Townsley et al. (2003) explained the diffuse soft X-ray emission
198: found in M17 in the context of strong shocks by 
199: fast stellar winds from young massive stars
200: \citep{Dyson1972,Castor1975,Weaver1977}.
201: Recent results on NGC 6334 by \cite{Ezoe2006} indicate that
202: the spectra of the diffuse emission varies from place to place; 
203: those in tenuous molecular cloud regions are soft and thermal
204: with temperatures of several keV, while those in dense cores 
205: exhibit harder continua with a photon index of $\Gamma\sim$1.
206: They also have shown that these thermal and non-thermal spectra of 
207: the diffuse X-ray emission in MSFRs, found as a mixture in the 
208: NGC 6334 case, may be generally explained 
209: by the stellar-wind shock model.
210: In spite of these observational progresses, even with {\it Chandra},
211: there remains an uncertainty how much undetected faint point sources 
212: contribute to the emission, because these MSFRs are relatively 
213: distant ($>1.4$kpc).
214: 
215: In order to unambiguously examine the diffuse X-ray emission 
216: in MSFRs as possible as we can, we analyzed archival {\it Chandra} 
217: data of NGC 2024. 
218: This region, known as the Flame nebula, is one of the nearest MSFRs 
219: ($D=415$ pc; \citealt{Anthony1982}), located in the Orion Nebula. 
220: It is an HII region considered to be illuminated by an O8V-B2V star 
221: IRS 2b \citep{Bik2003}. In the vicinity of IRS 2b, there are 
222: one early B star candidate associated
223: with an ultra compact HII region G206.543-16.347 and 
224: an infrared source IRS 2 \citep{Lenorzer2004}, 
225: and seven compact dust condensations named FIR 1-7, possibly 
226: massive protostars \citep{Mezger1988}. 
227: In addition, $\sim300$ low-mass 
228: ($\simlt2$\msolar) young stars have been found 
229: by near infrared observations \citep{Lada1991}. 
230: The estimated age of NGC 2024 ranges from 0.3 \citep{Meyer1996} 
231: to several Myr \citep{Comeron1996}.
232: A previous analysis of the same {\it Chandra} data on point sources 
233: has been published by \cite{Skinner2003}. In this paper, we focus on 
234: a search for the diffuse X-ray emission in this representative MSFR.
235: 
236: \section{Observation}
237: \label{sec:obs}
238: 
239: {\it Chandra} observed NGC 2024 on 2001 August 8-9 
240: using the ACIS--I for 21.9 hr.
241: We started with the level 1 event files in the same way
242: as \cite{Skinner2003}. Utilized analysis software versions 
243: for the standard data reduction are different; we used the 
244: CIAO (Chandra Interactive Analysis 
245: of Observations) software version 2.3 and the calibration 
246: data base version 2.18.
247: These new softwares allowed us to correct the 
248: data for the effect of the charge transfer inefficiency, 
249: while it was not possible in \cite{Skinner2003}. 
250: No background flares were seen during the observation with the
251: average count rate of the six ACIS chips of 9.1 ct s$^{-1}$.
252: We then excluded $>1.2$ times of the average rate.
253: This procedure excluded 2\% of the exposure time.
254: After these data screenings, the nominal exposure 
255: has become 75.3 ksec. 
256: 
257: We searched the data for diffuse X-ray emission, 
258: following to the analysis method of \cite{Ezoe2006}. 
259: We first created adaptively smoothed 
260: X-ray images in two energy bands, as shown in figures 
261: \ref{fig:image} (a) and (b).
262: We see a sign of an extended emission associated with the massive
263: star IRS 2b and its vicinity. The extended emission in 0.5--2 keV is 
264: strong in the north-west direction of IRS 2b, while that in 2--7 keV 
265: is elongated in the north-west to the south-east direction.
266: We then used the {\tt wavedetect} program to identify sources 
267: using images in three energy bands (0.5--2, 2--8 and 0.5--8 keV) 
268: independently, in order not to miss very soft or strongly 
269: absorbed sources. The significance criterion and wavelet scales
270: were set at $1\times10^{-6}$ and 1-16 pixels in multiples of $\sqrt{2}$.
271: In the ACIS--I field of view, 301 sources were detected. Among 
272: them, 28 sources are newly identified ones, not listed in 
273: \cite{Skinner2003}, due to our three-band searching method. 
274: 
275: We excluded all the detected point sources by creating a point source 
276: mask using the ``Chandra Ray Tracer''$\footnote{http://cxc.harvard.edu/chart/threads/index.html}$.
277: For each source, we defined a radius to include $\sim98$ \% 
278: of photons at the Al K$\alpha$-line energy (1.497 keV).
279: Then we excluded all these regions by applying the mask to 
280: the raw ACIS--I image, and created images of the residual 
281: emission using the CIAO tools {\tt dmfilth} and {\tt csmooth}. 
282: Figures \ref{fig:image} (c) and (d) show thus obtained images 
283: of the extended emission.
284: In order to evaluate the significance of this emission,
285: we defined a circular region named C1 (the large circle in the panel d
286: of figure \ref{fig:image}) with a radius of 4$^\prime=$0.5 pc.
287: The total area of the C1 is 43/39 arcmin$^2$ or 0.62/0.52 pc$^2$ before/after
288: excluding the area around point sources. The 0.5--7 keV count rate
289: of the C1 is $0.142\pm0.001$ cts s$^{-1}$, while that of the same region
290: in blank-sky data is $0.084\pm0.001$ cts s$^{-1}$. 
291: Errors are 1$\sigma$ statistical ones.
292: Hence, its residual count rate is  $0.057\pm0.001$ cts s$^{-1}$ 
293: or 4270$\pm$110 cts. It is thus clear that a highly significant 
294: extended emission is present in the central region of NGC 2024.
295: 
296: \section{Extended X-ray Emission}
297: \label{sec:extended}
298: 
299: In the presence of the significant excess emission, 
300: we immediately considered a possibility of the diffuse emission. 
301: To know its basic properties, we compared its background subtracted 
302: spectrum in figure \ref{fig:spec} (a) with that summed over 176 
303: point sources detected within the C1. The weighted ARF (ancillary 
304: response function) and RMF (response matrix function) were calculated
305: using the CIAO programs {\tt mkwarf} and {\tt mkwrmf}, respectively.
306: The {\tt apply\_acisabs} script was utilized when creating ARF files,
307: to correct them for the decrease in the ACIS quantum efficiency.
308: The background spectra are extracted from the same regions in
309: the blank-sky data for individual regions. 
310: From figure \ref{fig:spec} (a), we can see important features of 
311: the extended emission. 
312: First, it is about one order of magnitude fainter than the 
313: summed point sources. Second, the extended emission show a harder 
314: continuum in the 1--7 keV range. Third, a hint of an emission line is 
315: seen in 6--7 keV.
316: 
317: To know basic parameters of the extended emission, we then conducted
318: spectral fitting. We employed a simple power-law model with an interstellar 
319: absorption and a narrow Gaussian. Here and hereafter all quoted errors
320: in the spectral fitting refer to 90\% confidence levels unless otherwise noted.
321: The result is shown in figure \ref{fig:spec} (b), where table \ref{tbl:fit} 
322: lists the obtained parameters.
323: The fit was not acceptable with $\chi^2/\nu\sim1.5$ because of the excess 
324: around 0.5--1 keV. The line center energy indicates a He-like Fe K$_\alpha$ 
325: line from thermal plasma. Then, we consider an alternative ``leaky absorber'' 
326: condition; a single thermal emission component reaches us via two (or more)
327: paths with different absorptions. 
328: This situation is possible in the MSFR like NGC 2024 where the density of 
329: the molecular cloud varies from place to place.
330: We hence fitted the spectrum by a sum of two thermal components with 
331: independent absorptions, but with their temperatures and abundances together.
332: The result is shown in figure \ref{fig:spec} (c) and table \ref{tbl:fit}.
333: As a thin thermal plasma model, we utilized the 
334: APEC (astrophysical plasma emission code)$\footnote{http://hea-www.harvard.edu/APEC/}$
335: model. Then, the fitting result is improved from $\chi^2/\nu\sim1.5$ to 1.3, 
336: which is significant with 99.7\% confidence according to an F-test.
337: The soft excess is represented by the mildly absorbed high temperature emission. 
338: The best fit temperature is high 11 keV, consistent with the small photon index 
339: of $\Gamma\sim0.9$ in the power-law model fit.
340: The 0.5--7 keV flux obtained from the leaky absorption model is 1.1$\times10^{-12}$ \flux.
341: 
342: For comparison, we quantified the summed point-source spectrum.
343: In the same manner of the extended emission, the source spectrum, 
344: the background, ARF, and RMF files were obtained. 
345: The fit for a single thermal emission model with one absorption
346: was not acceptable with $\chi^2/\nu\sim7$ because
347: this model cannot represent both soft excess below $\sim2$ keV, 
348: similar to that in the extended emission case, and also various 
349: emission lines. Hence, we used the leaky absorption
350: model with free Ne, Mg, Si, S, Ar, Ca, and Fe abundances, in order 
351: to represent the data better.
352: We obtained results as shown in table \ref{tbl:fitsrc} and 
353: figure \ref{fig:specsrc}. The best-fit temperature of 4.4 keV 
354: is consistent with typical values of young low-mass stars 
355: (e.g., \citealt{Imanishi2001a}) and significantly lower 
356: than that of the extended emission.
357: 
358: In spite of these spectral analysis, 
359: We here must evaluate an effect of photons escaping from 
360: the summed point sources because the point sources are
361: far brighter than the extended emission.
362: In the same way as \cite{Ezoe2006}, we 
363: took into account the summed point source spectrum 
364: by multiplying the best-fit 
365: model for the summed point sources by the energy-dependent 
366: escaping-fraction curve, estimated by the ``{\it Chandra} Ray Tracer'' (ChaRT). 
367: We have found that this contributes $\sim40$\% to the extended emission.
368: we remove their 98\% photons at 2 keV.
369: After correcting the escaping photon effect, the 0.5--7 keV flux of the 
370: extended emission becomes 6.3$\times10^{-13}$ \flux, yielding an 
371: absorption-uncorrected luminosity of 1.3$\times10^{31}$ \ergs or
372: the surface brightness of 2$\times10^{31}$ \ergs pc$^{-2}$ at 
373: an assumed distance of 415 pc.
374: 
375: We refitted the C1 spectrum including the escaping photon effect. 
376: The result is shown in figure \ref{fig:spec} (d) and table \ref{tbl:fit}. 
377: The fitting result was again acceptable. Also, the escaping photon effect 
378: has a relatively little effect on the fitting parameters except the 
379: normalizations. All the parameters are consistent with the previous ones 
380: before including this effect within 90\% errors. 
381: The absorption-corrected X-ray luminosity is 2$\times10^{31}$ \ergs.
382: 
383: In addition, we also conducted the same spectral analysis to 
384: a C2 region, the soft X-ray clump seen in figure \ref{fig:image} (c).
385: The obtained photon index and temperature are similar to 
386: those of the C1 region within errors, except a lower 
387: absorption column density ($0.1\times10^{22}$ cm$^2$ 
388: when fitted with the leaky absorption model including 
389: the escaping photon effect). 
390: No sign of emission lines was seen.
391: Hence, the spectral hardness of the extended emission is 
392: considered to be common within the whole region, and the 
393: offset peak in the soft band map is simply a consequence 
394: of a slight reduced absorption.
395: 
396: \section{Luminosity Function}
397: \label{sec:lf}
398: 
399: Based on the surface brightness of the extended emission, 
400: we estimated contribution from unresolved faint point sources
401: to this emission, in order to know whether it can be explained
402: by faint sources that are individually undetectable. 
403: We followed the same way of \cite{Ezoe2006} in which we utilized
404: the luminosity function of the detected point sources.
405: We below utilize the X-ray surface brightness of the extended 
406: emission after subtracting the escape photons from the point sources, 
407: obtained in the last section. 
408: 
409: Figure \ref{fig:lf} shows the luminosity function 
410: of the 176 point sources in the C1. We derived
411: the absorption-uncorrected 0.5--7 keV luminosity $L$.
412: The X-ray flux of each point source is obtained 
413: by spectral fitting for a bright source ($\geq30$ net counts
414: or counts after subtracting the background), while by using 
415: a count-to-flux conversion factor derived from the 
416: summed point sources 1.6$\times10^{-11}$ \flux (net counts/s)$^{-1}$
417: for a fainter source.
418: 
419: The source number density increases toward lower luminosities 
420: and saturates below 10 net counts, corresponding to the completeness 
421: limit of this observation. We here estimated the limit from source 
422: number histograms as a function of the logarithm of net counts, 
423: and regarded the maximum of the histogram as the completeness limit.
424: When utilizing the above conversion factor, 
425: 10 net counts correspond to 4$\times$10$^{28}$ \ergs, in terms 
426: of the absorption-uncorrected 0.5--7 keV luminosity. 
427: This limit is one of the most lowest ones among those
428: in the past {\it Chandra} observations of MSFRs.
429: 
430: The solid line in figure \ref{fig:lf} indicates 
431: the necessary point source number, in order to 
432: explain the extended emission by point sources. 
433: The point number is clearly short, which supports 
434: that the extended emission is truly diffuse. 
435: Furthermore, even when we extrapolate the luminosity function 
436: toward a lower limit of 0 \ergs, using a linear function fitted in log-log space
437: from the completeness limit to $10^{30}$ \ergs, the estimated 
438: contribution of unresolved sources is at most $\sim10$\% of 
439: the extended emission. Based on these results, the extended 
440: emission of NGC 2024 can be considered as diffuse in nature.
441: 
442: \section{Discussion}
443: \label{sec:discuss}
444: 
445: We have found the diffuse X-ray emission in NGC 2024. 
446: Because of the proximity of NGC 2024 and the long exposure time, 
447: this discovery itself is one of the strongest supports for the existence 
448: of the diffuse emission in MSFRs, among the previous {\it Chandra} results.
449: Also, this result provides us with a new observational evidence that, 
450: even in a MSFR where only late O to early B stars exist, 
451: the diffuse X-ray emission can be observed if the sensitivity is enough high.
452: The spectral analysis suggests that the diffuse emission 
453: is dominated by the thermal emission. At the same time, since the continuum 
454: ($kT>8$ keV) of the diffuse emission is harder than typical spectra of 
455: young stars, a part of the emission may come from non-thermal origin. 
456: In NGC 6334, the soft-thermal and hard possibly non-thermal regions can 
457: be spatially distinguished \citep{Ezoe2006}.
458: In NGC 2024, the soft and hard regions may be co-spatial and hence 
459: be observed as a mixture of both thermal and non-thermal emission.
460: 
461: We then discuss whether the thermal or non-thermal interpretation 
462: is feasible in terms of energetics. First, if we assume that the 
463: whole diffuse emission is thermal, the total plasma energy $U$ 
464: will be $\sim10^{47}\eta^{0.5}$\ergs where $\eta$ is a filling factor,
465: from the equation (3) in \cite{Ezoe2006}.  
466: Using the X-ray luminosity $L_{\rm X}$ and $U$, 
467: the plasma cooling time is estimated as $U/L_{\rm X}\sim10^8\eta^{0.5}$ 
468: yr, which is far longer than the age of NGC 2024, 
469: from 0.3 \citep{Meyer1996} to several Myr \citep{Comeron1996}. 
470: Although the sound crossing time in a 10 keV plasma across the region 
471: of 0.5 pc in size is $\sim10^3$ yr and hence short, the plasma with
472: the estimated pressure of $\sim5\times10^7 \eta^{-0.5}$ K cm$^{-3}$, 
473: calculated from the equation (5) in \cite{Ezoe2006}, may be confined 
474: by the dense HII region known to exist around the molecular cloud 
475: dark lane (figure \ref{fig:image} c)  \citep{Subrahmanyan1997}
476: and also the strong magnetic field 
477: within the molecular cloud \citep{Crutcher1999}. 
478: Then, by dividing the total energy by the assumed age of NGC 2024 of 
479: 0.3 Myr, the average energy input is estimated as $10^{34}$ \ergs 
480: 
481: Second, we consider the non-thermal possibility.
482: The flat continuum ($\Gamma\sim1$ or $kT>8$ keV) of the emission strongly 
483: suggests the bremsstrahlung emission by 10 keV to several MeV electrons, 
484: rather than the synchrotron or inverse Compton emission \citep{Ezoe2006}. 
485: Since the Coulomb loss overwhelms the bremsstrahlung emission, if we assume 
486: that the diffuse emission is totally non-thermal, the necessary kinematic 
487: energy to be supplied is at least $\sim$10$^5$ times larger than the 
488: observed X-ray luminosity, $\sim$2$\times$10$^{36}$ \ergs. 
489: 
490: One of the most plausible energy sources for the diffuse emission is 
491: the shocks generated by fast stellar winds ($\sim2000$ km s$^{-1}$) 
492: from young massive stars. Its huge kinematic energy of the stellar 
493: winds can be easily converted via the strong shocks, among dense 
494: molecular clouds and HII regions, into the thermal (and non-thermal)
495: energy of particles in the surrounding gases.
496: This explanation has been proposed in M17 \citep{Townsley2003} 
497: and NGC 6334 \citep{Ezoe2006}.
498: In NGC 2024, there are at least one late O to early B star IRS 2b (O8V-B2V),
499: one early B star candidate IRS 2, 
500: and 7 possibly massive protostars. As shown in figure \ref{fig:image}, 
501: these massive stars are spatially associated with the diffuse emission. 
502: A typical kinematic luminosity of the stellar wind is 
503: $\sim10^{33-35}$ \ergs 
504: per one B2V-O8V star \citep{Howarth1989,Prinja1990,Panagia1973}. 
505: Therefore, the necessary energy supply of the thermal 
506: interpretation is explained if there is at least one 
507: massive star with a strong wind, comparable or stronger
508: than those of a typical B0.5V star.
509: The non-thermal interpretation is also possible if we consider 
510: all the 9 sources are massive stars and have strong winds. 
511: Wind-wind collisions may effectively increase the energy of a shock. 
512: Hence, the stellar wind scenario is possible from 
513: the viewpoint of the energetics.
514: 
515: %%Other observed parameters such as the plasma temperature and
516: %%the possible non-thermal component can be also explained by
517: %%the gas heating and particle acceleration in the stellar wind shocks.
518: %%Also, since the hot wind gas will expand since the energy supply of the
519: %%central OB star keeps constant, we can estimate the size 
520: %%of the expanding hot wind bubble as $\sim1$ pc, by assuming that 
521: %%the wind energy is equal to the displace energy of the cold gas 
522: %%surrounding the OB star. This estimated size is also consistent 
523: %%with the observed size. Hence, all these observed quantities 
524: %%support the stellar wind scenario.
525: 
526: YE is financially supported by 
527: the Japan Society for the Promotion of Science.
528: 
529: \begin{thebibliography}{}
530: 
531: \bibitem[\protect\astroncite{{Anthony-Twarog}}{1982}]{Anthony1982}
532: {Anthony-Twarog}, B.~J.: 1982,
533: \newblock {\em \aj} {\bf 87}, 1213
534: 
535: \bibitem[\protect\astroncite{{Bik} et~al.}{2003}]{Bik2003}
536: {Bik}, A., {Lenorzer}, A., {Kaper}, L., {Comer{\' o}n}, F., {Waters},
537:   L.~B.~F.~M., {de Koter}, A., and {Hanson}, M.~M.: 2003,
538: \newblock {\em \aap} {\bf 404}, 249
539: 
540: \bibitem[\protect\astroncite{{Castor} et~al.}{1975}]{Castor1975}
541: {Castor}, J., {McCray}, R., and {Weaver}, R.: 1975,
542: \newblock {\em \apjl} {\bf 200}, L107
543: 
544: \bibitem[\protect\astroncite{{Comeron} et~al.}{1996}]{Comeron1996}
545: {Comeron}, F., {Rieke}, G.~H., and {Rieke}, M.~J.: 1996,
546: \newblock {\em \apj} {\bf 473}, 294
547: 
548: \bibitem[\protect\astroncite{{Crutcher} et~al.}{1999}]{Crutcher1999}
549: {Crutcher}, R.~M. and {Roberts}, D.~A. and {Troland}, T.~H. and 
550: 	{Goss}, W.~M.: 1999,
551: \newblock {\em \apj} {\bf 515}, 275
552: 
553: \bibitem[\protect\astroncite{{Dyson} and {de Vries}}{1972}]{Dyson1972}
554: {Dyson}, J.~E. and {de Vries}, J.: 1972,
555: \newblock {\em \aap} {\bf 20}, 223
556: 
557: \bibitem[\protect\astroncite{{Ezoe} et~al.}{2006}]{Ezoe2006}
558: {Ezoe}, Y., {Kokubun}, M., {Makishima}, K., {Sekimoto}, Y., and {Matsuzaki},
559:   K.: 2006,
560: \newblock {\em \apj} {\bf 638}, 860
561: 
562: \bibitem[\protect\astroncite{{Giacconi} et~al.}{2001}]{Giacconi2001}
563: {Giacconi}, R., {Rosati}, P., {Tozzi}, P., {Nonino}, M., {Hasinger}, G.,
564:   {Norman}, C., {Bergeron}, J., {Borgani}, S., {Gilli}, R., {Gilmozzi}, R., and
565:   {Zheng}, W.: 2001,
566: \newblock {\em \apj} {\bf 551}, 624
567: 
568: \bibitem[\protect\astroncite{{Howarth} et~al.}{1989}]{Howarth1989}
569: {Howarth}, I.~D. and {Prinja}, R.~K.: 1989,
570: \newblock {\em \apjs} {\bf 69}, 527
571: 
572: \bibitem[\protect\astroncite{{Imanishi} et~al.}{2001}]{Imanishi2001a}
573: {Imanishi}, K., {Koyama}, K., and {Tsuboi}, Y.: 2001,
574: \newblock {\em \apj} {\bf 557}, 747
575: 
576: \bibitem[\protect\astroncite{{Lada}}{1991}]{Lada1991}
577: {Lada}, C.: 1991,
578: \newblock {\em {The Physics of Star Formation and Early Evolution}},
579: \newblock Advanced Sci. Ser., ed. CJ Lada \& ND Kylafis, 1991.\ 329 p.
580: 
581: \bibitem[\protect\astroncite{{Lenorzer} et~al.}{2004}]{Lenorzer2004}
582: {Lenorzer}, A., {Bik}, A., {de Koter}, A., {Kurtz}, S.~E., {Waters},
583:   L.~B.~F.~M., {Kaper}, L., {Jones}, C.~E., and {Geballe}, T.~R.: 2004,
584: \newblock {\em \aap} {\bf 414}, 245
585: 
586: \bibitem[\protect\astroncite{{Meyer}}{1996}]{Meyer1996}
587: {Meyer}, M.~R.: 1996,
588: \newblock {\em \pasp} {\bf 108}, 380
589: 
590: \bibitem[\protect\astroncite{{Mezger} et~al.}{1988}]{Mezger1988}
591: {Mezger}, P.~G., {Chini}, R., {Kreysa}, E., {Wink}, J.~E., and {Salter}, C.~J.:
592:   1988,
593: \newblock {\em \aap} {\bf 191}, 44
594: 
595: \bibitem[\protect\astroncite{{Moffat} et~al.}{2002}]{Moffat2002}
596: {Moffat}, A.~F.~J., {Corcoran}, M.~F., {Stevens}, I.~R., {Skalkowski}, G.,
597:   {Marchenko}, S.~V., {M{\" u}cke}, A., {Ptak}, A., {Koribalski}, B.~S.,
598:   {Brenneman}, L., {Mushotzky}, R., {Pittard}, J.~M., {Pollock}, A.~M.~T., and
599:   {Brandner}, W.: 2002,
600: \newblock {\em \apj} {\bf 573}, 191
601: 
602: \bibitem[\protect\astroncite{{Panagia}}{1973}]{Panagia1973}
603: {Panagia}, N.: 1973,
604: \newblock {\em \aj} {\bf 78}, 929
605: 
606: \bibitem[\protect\astroncite{{Prinja}}{1990}]{Prinja1990}
607: {Prinja}, R.~K. and {Barlow}, M.~J. and {Howarth}, I.~D.: 1990,
608: \newblock {\em \apj} {\bf 361}, 607
609: 
610: \bibitem[\protect\astroncite{{Skinner} et~al.}{2003}]{Skinner2003}
611: {Skinner}, S., {Gagne}, M., and {Belzer}, E.: 2003,
612: \newblock {\em \apj} {\bf 598}, 375
613: 
614: \bibitem[\protect\astroncite{{Subrahmanyan} et~al.}{1997}]{Subrahmanyan1997}
615: {Subrahmanyan}, R. and {Goss}, W.~M. and {Megeath}, S.~T. and 
616: 	{Barnes}, P.~J.: 1997,
617: \newblock {\em \apj} {\bf 290}, 431
618: 
619: \bibitem[\protect\astroncite{{Townsley} et~al.}{2003}]{Townsley2003}
620: {Townsley}, L.~K., {Feigelson}, E.~D., {Montmerle}, T., {Broos}, P.~S., {Chu},
621:   Y., and {Garmire}, G.~P.: 2003,
622: \newblock {\em \apj} {\bf 593}, 874
623: 
624: \bibitem[\protect\astroncite{{Townsley} et~al.}{2004}]{Townsley2004}
625: {Townsley}, L.~K., {Feigelson}, E.~D., {Montmerle}, T., {Broos}, P.~S., {Chu},
626:   Y., and {Garmire}, G.~P.: 2004,
627: \newblock {\em astro-ph/0406349}
628: 
629: \bibitem[\protect\astroncite{{Wang} et~al.}{2002}]{Wang2002}
630: {Wang}, Q.~D., {Gotthelf}, E.~V., and {Lang}, C.~C.: 2002,
631: \newblock {\em \nat} {\bf 415}, 148
632: 
633: \bibitem[\protect\astroncite{{Weaver} et~al.}{1977}]{Weaver1977}
634: {Weaver}, R., {McCray}, R., {Castor}, J., {Shapiro}, P., and {Moore}, R.: 1977,
635: \newblock {\em \apj} {\bf 218}, 377
636: 
637: \bibitem[\protect\astroncite{{Wolk} et~al.}{2002}]{Wolk2002}
638: {Wolk}, S.~J., {Bourke}, T.~L., {Smith}, R.~K., {Spitzbart}, B., and {Alves},
639:   J.: 2002,
640: \newblock {\em \apjl} {\bf 580}, L161
641: 
642: \bibitem[\protect\astroncite{{Yusef-Zadeh} et~al.}{2002}]{Yusef-Zadeh2002}
643: {Yusef-Zadeh}, F., {Law}, C., {Wardle}, M., {Wang}, Q.~D., {Fruscione}, A.,
644:   {Lang}, C.~C., and {Cotera}, A.: 2002,
645: \newblock {\em \apj} {\bf 570}, 665
646: 
647: \end{thebibliography}
648: 
649: \clearpage
650: 
651: \begin{figure}
652: \begin{center}
653: \includegraphics[width=0.7\textwidth,clip,angle=-90]{f1.eps}
654: \caption{Adaptively smoothed X-ray images of NGC 2024 taken 
655:          with the {\it Chandra} ACIS--I. 
656:          Panels (a) and (b) correspond to 0.5--2 keV and 
657:          2--7 keV band images before applying the point source mask, 
658:          respectively, while panels (c) and (d) are point-source
659:          excluded contours in 0.5--2 and 2--7 keV bands overlaid
660:          on the optical DSS image. 
661:          Co-ordinates are the J2000. All the images are 
662:          corrected for the exposure and vignetting, but the background 
663:          is not subtracted. 
664: 	 Boxes show the ACIS--I fields of view, while large circles (C1 and C2)
665:          are regions utilized in our spectral analysis.
666:          Small circles are positions of IRS 2b, IRS2 and FIR 1-7.
667:          Because IRS 2 is located just 5$^{\prime\prime}$ south-east of IRS2b,
668:          these two circles are overlapped. 
669:          The color intensity is plotted logarithmically 
670: 	 from 5$\times10^{-10}$ to 5$\times10^{-6}$
671:          counts s$^{-1}$ pixel$^{-1}$ cm$^{-2}$ in panel (a),
672:          while from 7$\times10^{-10}$ to 7$\times10^{-6}$ in panel (b).
673: 	 Also the contours are plotted logarithmically from 
674: 	 7.5$\times10^{-10}$ to 1.0$\times10^{-9}$
675: 	 counts s$^{-1}$ pixel$^{-1}$ cm$^{-2}$ in panel (c),
676:          while from 1.3$\times10^{-9}$ to 4.0$\times10^{-9}$ in panel (d).
677:          }
678: \label{fig:image}
679: \end{center}
680: \end{figure}
681: 
682: \clearpage
683: 
684: \begin{figure}
685: \centerline{
686: \includegraphics[width=0.35\textwidth,clip,angle=-90]{f2a.eps}
687: \hspace*{0.02\textwidth}
688: \includegraphics[width=0.35\textwidth,clip,angle=-90]{f2b.eps}
689: }
690: \bigskip
691: \centerline{                                                         
692: \includegraphics[width=0.35\textwidth,clip,angle=-90]{f2c.eps}
693: \hspace*{0.02\textwidth}
694: \includegraphics[width=0.35\textwidth,clip,angle=-90]{f2d.eps}
695: }                                                                    
696: \caption{(a) ACIS spectrum of the C1 region compared with that 
697:          summed over 176 point sources.
698:          (b) The C1 spectrum fitted with a power-law plus a narrow
699:          Gaussian model.
700:          (c) The same as panel b, but for a single temperature plasma
701:          model suffering two absorption column densities (solid and dotted lines).
702:          (d) The same as panel c, but including the escaped photons from
703:          the summed point sources (red lines).
704:          }
705: \label{fig:spec}
706: \end{figure}
707: 
708: \clearpage
709: 
710: \begin{figure}
711: \centerline{
712: \includegraphics[width=0.5\textwidth,clip,angle=-90]{f3.eps}
713: }
714: \caption{The summed point-source spectrum in the C1 region, fitted
715:          with the leaky absorption model (solid and dotted lines).
716:          }
717: \label{fig:specsrc}
718: \end{figure}
719: 
720: \clearpage
721: 
722: \begin{figure}
723: \centerline{
724: \includegraphics[width=0.5\textwidth,clip,angle=-90]{f4.eps}
725: }
726: \caption{ Luminosity function of the point sources within the C1.
727:           The vertical axis is the column density of the source number,
728:           while the horizontal axis absorption-uncorrected 0.5--7 keV
729:           luminosity. Errors are 1$\sigma$ Poisson. The thick line 
730:           indicates the number of putative point sources at a given
731:           flux that can account for the observed total surface brightness
732:           of the extended emission. Three short vertical lines are the
733:           typical X-ray fluxes of point sources with 10, 20 and 30 net 
734:           counts. The dashed line indicates the expected number of 
735:           extragalactic sources, estimated from \cite{Giacconi2001}.
736:          }
737: \label{fig:lf}
738: \end{figure}
739: 
740: \clearpage
741: 
742: \begin{table}
743: \caption{Parameters of the spectral fitting to the extended emission 
744:          in the C1 region.}
745: \label{tbl:fit}
746: {\scriptsize
747: \begin{tabular}{lccccccccccccc} \hline\hline
748:        Model$^{\rm a}$  & $N_{\rm H1}$$^{\rm b}$   & $N_{\rm H2}$$^{\rm b}$ 
749:        & $\Gamma$ or $kT^{\rm c}$    & $Z^{\rm d}$ 
750:        & Norm.1$^{\rm e}$  & Norm.2$^{\rm e}$           
751:        & $E_{\rm C}^{\rm f}$         & Norm.$_{\rm line}^{\rm g}$ 
752:        & $F_{\rm X}^{\rm h}$         & $L_{\rm X}^{\rm i}$ 
753:        & $\chi^2$/d.o.f. \\\hline
754: 
755:        po+ga & 1.3$^{+0.5}_{-0.3}$ & $-$
756:        & 0.94$^{+0.32}_{-0.24}$ & $-$
757:        & 1.2$^{+0.6}_{-0.5}$ & $-$
758:        & 6.7$^{+0.1}_{-0.2}$    & 6.3$^{+4.6}_{-4.0}$ 
759:        & 1.1 & 2.9
760:        & 99.2/65 \\
761: 
762:        leaky abs.1  & 0.62$(<1.1)$ & 4.0$^{+1.4}_{-1.5}$
763:        & 11$^{+8.8}_{-3.5}$     & 1.3$^{+1.6}_{-0.8}$    
764:        & 2.2$^{+1.6}_{-1.7}$ & 9.2$^{+2.7}_{-1.6}$ 
765:        & $-$ & $-$
766:        & 1.1 & 4.0 
767:        & 85.9/64 \\
768: 
769:        leaky abs.2  & 0.21$(<1.1)$ & 3.3$^{+7.7}_{-1.0}$ 
770:        & 11($>$7.6)       & 2.5($>$1.0)
771:        & 0.73$^{+2.27}_{-0.43}$ & 4.4$^{+1.6}_{-1.4}$ 
772:        & $-$ & $-$
773:        & 0.63 & 2.2
774:        & 84.2/64 \\\hline
775: 
776: \end{tabular}
777: }
778: {\noindent
779: $^{\rm a}$ Fitting models. The po+ga, leaky abs.1 and leaky abs.2 indicate 
780:      the power-law plus a narrow Gaussian model, the leaky absorption model
781:      without and with the escaping photon effect from the point sources, respectively. \\
782: $^{\rm b}$ Interstellar absorption, with $N_{\rm H}$ being the hydrogen column density in 10$^{22}$ cm$^{-2}$. 
783:      In the case of the leaky absorption model, two column densities are given.\\
784: $^{\rm c}$ $\Gamma$ is the photon index in the power-law model, while $kT$ is a plasma temperature in keV.\\
785: $^{\rm d}$ Metal abundance in solar, used in the leaky absorption model. \\
786: $^{\rm e}$ Normalization is photon flux at 1 keV in 10$^{-4}$ photons cm$^{-2}$ s$^{-1}$ in the power-law model,
787:      and 10$^{-18}$/4$\pi D^2$ $EM$ in the leaky absorption model, where $D$ is a distance to NGC 2024 and 
788:      $EM$ is an emission measure in cm$^{-3}$.\\
789: $^{\rm f}$ A line center energy in keV.\\
790: $^{\rm g}$ Line intensity in 10$^{-6}$ photons cm$^{-2}$ s$^{-1}$. \\
791: $^{\rm h}$ The 0.5--7 keV flux in 10$^{-12}$ \flux. \\
792: $^{\rm i}$ The absorption-corrected 0.5--7 keV luminosity in 10$^{31}$ \ergs.
793: }
794: \end{table}
795: 
796: \clearpage
797: 
798: \begin{table}
799: \caption{Parameters of the spectral fitting to the summed point sources in the C1,
800:          derived by the leaky absorption model$^{\rm a}$.}
801: \label{tbl:fitsrc}
802: \begin{center}
803: \begin{tabular}{cccccccccccccc} \hline\hline
804: param & leaky abs. model \\\hline
805: $N_{\rm H1}$ & 0.55$\pm0.09$  \\
806: $N_{\rm H2}$ & 2.7$^{+0.3}_{-0.2}$  \\
807: $kT$         & 4.4$^{+0.3}_{-0.4}$  \\
808: $Z_{\rm Ne}$$^{\rm b}$ & 2.9$(>1.2)$  \\
809: $Z_{\rm Mg}$$^{\rm b}$ & 1.1$\pm0.8$  \\
810: $Z_{\rm Si}$$^{\rm b}$ & 0.06$(<0.32)$  \\
811: $Z_{\rm S} $$^{\rm b}$  & 0.93$^{+0.37}_{-0.36}$  \\
812: $Z_{\rm Ar}$$^{\rm b}$ & 1.2$^{+0.7}_{-0.5}$  \\
813: $Z_{\rm Ca}$$^{\rm b}$ & 0.59$(<1.2)$  \\
814: $Z_{\rm Fe}$$^{\rm b}$ & 0.23$^{+0.04}_{-0.03}$  \\
815: Norm.1       & 17$^{+5}_{-4}$  \\
816: Norm.2       & 130$\pm10$  \\
817: $F_{\rm X}$  & 9.5  \\ 
818: $L_{\rm X}$  & 43  \\
819: $\chi^2$/d.o.f. & 292.1/157\\\hline
820: \end{tabular}
821: \end{center}
822: {\noindent
823: $^{\rm a}$ Notations and symbols are the same as the leaky absorption model in table \ref{tbl:fit} 
824:            except abundances.\\
825: $^{\rm b}$ Metal abundances of individual elements in solar, while those of other elements
826:            are fixed at 1 solar.\\
827: }
828: \end{table}
829: 
830: \end{document}
831: 
832: %%
833: %% End of file.
834: