astro-ph0309271/ms.tex
1: \documentclass{aastex}
2: \usepackage{emulateapj5,times,mathptm}
3: \singlespace
4: \slugcomment{Accepted for the publication in the {\it Astrophysical Journal 
5: Letters}}
6: \pagenumbering{arabic}
7: \newcommand{\NH}{\mbox{$N_{\rm H}$}}
8: \newcommand{\hii}{H{\small II} }
9: \def\fdg{\hbox{$.\!\!^\circ$}}
10: \def\farcm{\hbox{$.\mkern-4mu^\prime$}}
11: \def\farcs{\hbox{$.\mkern-4mu^{\prime\prime}$}}
12: \def\fsecs{\hbox{$.\mkern-4mu^s$}}
13: \def\la{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$<$}}}}
14: \def\ga{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$>$}}}}
15: 
16: \shortauthors{Park}
17: \shorttitle{0103-72.6}
18: 
19: \begin{document}
20: 
21: \title{0103$-$72.6: A New Oxygen-Rich Supernova Remnant 
22: in the Small Magellanic Cloud}
23: 
24: \author{Sangwook Park\altaffilmark{1}, John P. Hughes\altaffilmark{2}, 
25: David N. Burrows\altaffilmark{1}, Patrick O. Slane\altaffilmark{3}, 
26: John A. Nousek\altaffilmark{1}, and Gordon P. Garmire\altaffilmark{1}}
27: 
28: \altaffiltext{1}{Department of Astronomy and Astrophysics, Pennsylvania State
29: University, 525 Davey Laboratory, University Park, PA. 16802; 
30: park@astro.psu.edu}
31: \altaffiltext{2}{Department of Physics and Astronomy, Rutgers University,
32: 136 Frelinghuysen Road, Piscataway, NJ. 08854-8019}
33: \altaffiltext{3}{Harvard-Smithsonian Center for Astrophysics, 60 Garden Street,
34: Cambridge, MA. 02138}
35: 
36: \begin{abstract}
37: 
38: 0103$-$72.6, the second brightest X-ray supernova remnant (SNR) 
39: in the Small Magellanic Cloud (SMC), has been observed with the 
40: {\it Chandra X-Ray Observatory}. Our {\it Chandra} observation 
41: unambiguously resolves the X-ray emission into a nearly complete, 
42: remarkably circular shell surrounding bright clumpy emission in 
43: the center of the remnant. The observed X-ray spectrum for the 
44: central region is evidently dominated by emission from reverse 
45: shock-heated metal-rich ejecta. Elemental abundances in this 
46: ejecta material are particularly enhanced in oxygen and neon, 
47: while less prominent in the heavier elements Si, S, and Fe. 
48: We thus propose that 0103$-$72.6 is a new ``oxygen-rich'' SNR,
49: making it only the second member of the class in the SMC.
50: The outer shell is the limb-brightened, soft X-ray emission 
51: from the swept-up SMC interstellar medium. The presence of O-rich 
52: ejecta and the SNR's location within an H{\small II} region attest 
53: to a massive star core-collapse origin for 0103$-$72.6. 
54: The elemental abundance ratios derived from the ejecta suggest an 
55: $\sim$18~M$_{\odot}$ progenitor star.
56: 
57: \end{abstract}
58: 
59: \keywords {ISM: individual (0103$-$72.6) --- supernova remnants --- 
60: X-rays: ISM}
61: 
62: \section {\label {sec:intro} INTRODUCTION}
63: 
64: Identifying the explosion type of a supernova remnant (SNR), 
65: largely between the thermonuclear (Type Ia) and the core-collapse 
66: (Type II or Ib/Ic) explosions, is difficult, yet important for 
67: understanding how heavy elements are synthesized and distributed 
68: into interstellar space. A small number of SNRs are known
69: to exhibit emission from SN ejecta material which is enhanced in 
70: O, Ne, and Mg. Based on the nucleosynthesis products of such
71: explosions, these ``oxygen-rich'' SNRs result from the core-collapse 
72: of massive progenitor stars. The O-rich SNRs thus provide a useful 
73: test-ground for the studies of SN nucleosynthesis. The number of 
74: known O-rich SNRs is small, however: there are three in the Galaxy 
75: (Cassiopeia A, Puppis A, and G292.0$+$1.8), two (N132D and 
76: 0540$-$69.3) in the Large Magellanic Cloud (LMC), and merely one 
77: (0102.2$-$7219) in the Small Magellanic Cloud (SMC). Although these 
78: SNRs are all classified as O-rich SNRs by virtue of the presence of 
79: ejecta material enriched in O, the nature of the individual objects 
80: varies significantly from one to another. The detection of more such 
81: SNRs is thus required for unbiased, systematic investigations of 
82: this interesting class of SNRs.
83: 
84: 0103$-$72.6 is the second brightest X-ray SNR in the SMC 
85: \citep{seward81,haberl00}. The X-ray images taken with the 
86: {\it ROSAT}/HRI showed a centrally-peaked morphology and a 
87: faint outer shell with an $\sim$3$'$ diameter \citep{yoko02}. 
88: The radio data indicated only a faint shell-like feature roughly
89: along the X-ray shell \citep{mills82,yoko02}. In the optical band, 
90: 0103$-$72.6 is very faint, probably elliptical, and ring-like with 
91: nearly the same angular size as the X-ray and the radio remnant 
92: \citep{math83}. Although the large angular size was suggestive of
93: a relatively old age ($\sim$10$^4$ yr), the {\it ASCA} data 
94: suggested elevated metal abundances \citep{yoko02}.
95: The chemical compositions of the overabundant material were
96: consistent with a core-collapse Type II SN nucleosynthesis model. 
97: This result was also supported by the SNR's location within an 
98: H{\small II} region (DEM S125) (e.g., Filipovi\'c et al. 1998 and 
99: references therein), which is consistent with a massive progenitor.
100:  
101: We report here on the results from our observation of 0103$-$72.6
102: with the {\it Chandra X-Ray Observatory}. The high resolution
103: {\it Chandra} data unambiguously resolve the SNR into a 
104: limb-brightened circular shell surrounding bright, clumpy central 
105: emission. The central regions are enriched in O and Ne whereas 
106: the outer shell is dominated by emission from the swept-up SMC 
107: ISM with low elemental abundances. Based on these results, we 
108: propose that 0103$-$72.6 is a new O-rich SNR in the SMC.
109: 
110: \section{\label{sec:obs} OBSERVATION \& DATA REDUCTION}
111: 
112: We observed SNR 0103$-$72.6 with the Advanced CCD Imaging
113: Spectrometer (ACIS) on board the {\it Chandra X-Ray Observatory}
114: on 2002 August 27 as part of the Guaranteed Time Observation
115: program. The ACIS-S3 chip, with  very faint (VFAINT) mode, was 
116: chosen. We screened the raw data in order to utilize the 
117: VFAINT mode and then filtered by the flight timeline filter. We 
118: corrected the charge transfer inefficiency (CTI; Townsley 
119: et al. 2000) with the methods developed by Townsley et al. (2002a).
120: The data were then screened by the standard filtering with
121: status and grade. We inspected the overall lightcurve for
122: any background flares. We found no significant variations of
123: the lightcurve for the $\sim$3$'$ diameter source region.
124: After this data reduction, the effective exposure was $\sim$49
125: ks with $\sim$45000 photons in the 0.3$-$3 keV band. 
126: The overall spectrum of 0103$-$72.6 is very soft and {\it all}
127: source photons are detected at $E$ $<$ 3 keV. 
128: 
129: \section{\label{sec:image} X-RAY IMAGES}
130: 
131: Figure~\ref{fig:fig1} is a ``true-color'' image of SNR 0103$-$72.6.
132: The high resolution ACIS images clearly resolve the bright central
133: emission and the limb-brightened outer shell. The outer shell is
134: remarkably circular and dominated by soft X-ray emission, indicated
135: in red. The morphological and spectral features of the circumferential 
136: shell are typical of X-ray emission from ISM swept-up by the 
137: blast wave shock front. In contrast, the central regions are bright 
138: in all colors and show a complex mixture of X-ray emitting knots. 
139: These features are qualitatively consistent with a Sedov picture. 
140: 
141: We explored the angular distributions of the X-ray line emission 
142: by constructing {\it equivalent width} (EW) images for the detected 
143: elemental species, following the method of Park et al. (2002). 
144: We generated EW images for O and Ne by selecting photons 
145: around the broad line features. We present the O EW image in
146: Figure~\ref{fig:fig2}. The O EW image clearly reveals
147: enhancements in O line emission in the central regions of 
148: 0103$-$72.6. In contrast, the circumferential shell is fairly 
149: featureless in the O EW image. These overall features are 
150: also consistent in the Ne EW image. These EW features support our 
151: conjecture regarding the overall characteristics of the SNR based 
152: on the color images: i.e., the central regions are emission from 
153: metal-rich ejecta heated by the reverse shock, and the outer shell 
154: is the emission from the forward shock sweeping through the 
155: ambient ISM.
156: 
157: \section{\label{sec:spec} X-ray Spectra} 
158: 
159: The high resolution ACIS images allow us to perform 
160: a spatially-resolved spectral analysis of 0103$-$72.6.
161: For the spectral analysis, we use detector response matrices,
162: as generated by Townsley et al. (2002b), appropriate for our 
163: CTI-corrected data. The low energy ($E$ $\la$ 1 keV) quantum 
164: efficiency (QE) of the ACIS has degraded because of molecular 
165: contamination on the optical blocking filter. We corrected 
166: this time-dependent QE degradation by modifying the ancillary 
167: response function for each extracted spectrum, utilizing 
168: the IDL ACISABS software\footnote{For the discussion on this 
169: instrumental issue, see 
170: http://cxc.harvard.edu/cal/Acis/Cal\_prods/qeDeg/index.html.
171: The software was developed by George Chartas and is available at
172: http://www.astro.psu.edu/users/chartas/xcontdir/xcont.html.}.
173: 
174: The presence of the limb-brightened shell, which is remarkably
175: circular around the centrally-bright emission, indicates a 
176: significant contribution from the foreground/background swept-up 
177: ISM to the central excess emission. In order to characterize this 
178: projected emission from the heated ISM, we first extracted spectra 
179: from the bright parts of the limb-brightened circumferential 
180: shell (``SW shell'' and ``SE shell'' regions in 
181: Figure~\ref{fig:fig3}). This outer shell is most likely 
182: dominated by emission from the blast wave shock front. 
183: We thus fitted these spectra simultaneously with a nonequilibrium 
184: ionization (NEI) plane-parallel shock model \citep{bor01} 
185: (Figure~\ref{fig:fig4}). Only the normalization was varied 
186: freely for each individual regional spectrum. These outer shell 
187: spectra can be best-fitted with $kT$ $\sim$ 0.25$-$0.30~keV, 
188: depending on the implied foreground column, which is poorly 
189: constrained ($N_H$ $<$ 1.2 $\times$ 10$^{21}$ cm$^{-2}$). The 
190: fitted metal abundances are low ($\sim$0.1 solar), which is in 
191: agreement with typical SMC ISM \citep{russell92}. 
192: 
193: We then extracted spectra from the east and the west side of the 
194: central excess emission, {\it interior} to the outer shell (i.e., 
195: ``east'' and ``west'' regions in Figure~\ref{fig:fig3}). We 
196: simultaneously fitted these spectra with an NEI Sedov model 
197: assuming that the electron temperature immediately behind the shock 
198: is $kT$~$\sim$~0.25$-$0.3~keV as obtained from the outer shell 
199: spectrum. Only the normalizations were allowed to vary independently. 
200: While the Sedov model corresponds to the full temperature and density 
201: distribution of the swept-up component of an entire SNR, we use this 
202: model as a first-order approximation to account for the expected 
203: temperature distribution for the plasma in the SNR's interior. The 
204: goal of these fits is to describe the projected blast wave emission 
205: seen in the central region in combination with the ejecta component.
206: These spectra can be described by $kT$~$\sim$~0.26~keV 
207: and low SMC-like abundances of 0.12$^{+0.08}_{-0.04}$ solar 
208: (Figure~\ref{fig:fig4}; Table~\ref{tbl:tab1}). 
209: 
210: The central region spectrum has been extracted from the ``center'' 
211: region (Figure~\ref{fig:fig3}). We fitted this spectrum with a
212: two-component model: i.e., a Sedov component for the projected
213: emission from the swept-up ISM, and a plane-shock model for the
214: central ejecta component (Figure~\ref{fig:fig5}). We fixed the 
215: Sedov component parameters at the best-fit values obtained above. 
216: This Sedov component for the projected emission from the swept-up 
217: ISM comprises $\sim$15\% of the observed flux for 
218: the ``center'' region. The best-fit ejecta electron temperature 
219: ($kT$~$\sim$~0.53~keV) is higher than that of the blast wave shock 
220: front as measured from the outer shell (Table~\ref{tbl:tab1}). 
221: The elemental abundances for the ejecta component were varied freely 
222: for O, Ne, Mg, Si and Fe while others are fixed at the SMC values 
223: \citep{russell92}. The fitted abundances are also higher than those 
224: of the SMC: particularly, O and Ne abundances are 4$-$6 times higher 
225: than typical SMC abundances. Mg and Si are moderately elevated 
226: ($\sim$1.5 times the SMC), and Fe abundance ($\sim$0.3 times the SMC) 
227: is very low. The S abundance is unconstrained with the current
228: data and thus fixed at the SMC abundance. The improvement made by 
229: fitting the S abundance is statistically negligible.
230: 
231: \section{\label{sec:disc} DISCUSSION}
232: 
233: The X-ray spectrum from the bright central region of 0103$-$72.6 
234: indicates the presence of ejecta material which is particularly
235: enriched in O and Ne. We thus propose that 0103$-$72.6 is a new
236: O-rich SNR, based on the results of our X-ray spectral 
237: analysis. 0103$-$72.6 is then the second member of this class
238: in the SMC in addition to SNR 0102.2$-$7219 \citep{dopita84}. 
239: The O- and Ne-dominated nature of the ejecta of 0103$-$72.6,
240: with the relative absence of high-Z species, particularly Fe, is 
241: similar to that of 0102.2$-$7219 \citep{blair00,sasaki01} and the 
242: Galactic O-rich SNR G292.0+1.8 \citep{park03a}. This abundance 
243: distribution however differs from that of Cas A, which shows 
244: significant Si-, S-, and Fe-rich ejecta \citep{hughes00a}. 
245: The physical location within an H{\small II} region and the
246: detected O-rich nature are typical signatures of a core-collapse
247: origin for 0103$-$72.6. We thus compare the measured elemental
248: abundances of 0103$-$72.6 with a Type II SN nucleosynthesis model
249: (Table~\ref{tbl:tab2}). Although the measured elemental abundances
250: have systematic uncertainties depending on the embedded contribution 
251: from the background swept-up ISM component, the quoted abundance 
252: ratios are nearly independent of these uncertainties. The 
253: measured abundance ratios are in plausible agreement with a Type 
254: II nucleosynthesis model for an 18~M$_{\odot}$ progenitor star. 
255: 
256: Assuming that 0103$-$72.6 is in the Sedov phase, we may derive 
257: the SNR parameters by taking advantage of the spatial information 
258: obtained by the high-resolution ACIS images.
259: Based on the best-fit volume emission measure 
260: ($EM$~=~5.4~$\times$~10$^{58}$~cm$^{-3}$ by combining four fitted 
261: shell regions; Table~\ref{tbl:tab1}), we derive the postshock 
262: electron density, $n_e$~=~0.79 cm$^{-3}$ for the blast wave shock 
263: front. In this estimation, we have assumed that the average outer 
264: radius of the SNR is $R$ $\sim$ 85$^{\prime\prime}$ (corresponding
265: to $\sim$24.7 pc at $d$ = 60 kpc) and that the blast wave shell 
266: is $\sim$20$^{\prime\prime}$ thick, based on the ACIS image.
267: The average depth through our shell regions is then $\sim$23 pc which 
268: implies a total volume for the four shell regions of $\sim$1 $\times$ 
269: 10$^{59}$ cm$^3$. The pre-shock hydrogen density is
270: then $n_0$ $\sim$ 0.16 cm$^{-3}$ assuming a factor of 4 density
271: compression behind shock and a mean charge state of $n_e$ = 1.2~$n_H$. 
272: Assuming uniform density for the ambient ISM, we estimate the mass 
273: swept-up by the blast wave to be $M_{swept}$~$\sim$~570 M$_{\odot}$.
274: The best-fit electron temperature of the shock front ($kT$ = 0.3 keV) 
275: corresponds to a shock velocity of $v$ $\sim$ 480 km s$^{-1}$, 
276: under the assumption of electron-ion temperature equilibration.
277: We then derive the SNR age of $t$ $\sim$ 1.8 $\times$ 10$^4$ yr
278: and the SN explosion energy of $E_0$ $\sim$ 1.2 $\times$ 10$^{51}$
279: ergs based on the Sedov solution. 0103$-$72.6 thus appears to be
280: the oldest known example of an O-rich SNR.
281: The detection of significant amounts of ejecta material in 
282: 0103$-$72.6 adds yet another such an example of the growing number 
283: of middle-aged SNRs with observable ejecta (e.g., Park et al. 2003b 
284: and references therein). 
285: 
286: We make a brief morphological comparison between the two SMC O-rich
287: SNRs, 0103$-$72.6 and 0102.2$-$7219. 0102.2$-$7219 is a young 
288: SNR ($\sim$10$^3$ yr) with a small angular size of 
289: $\sim$44$^{\prime\prime}$ diameter, and the X-ray morphology shows
290: a bright ring-like structure embedded in a relatively faint 
291: diffuse emission extending beyond the bright ring \citep{gaetz00}. 
292: This morphology was interpreted as the reverse shock front heating 
293: the central O-rich ejecta material (i.e., the sharp inner boundary 
294: of the bright ring) and the blast wave shock front propagating into 
295: the ambient ISM (i.e., the outer boundary of the faint extended 
296: diffuse emission) \citep{gaetz00}. In comparison, as we have just 
297: determined, 0103$-$72.6 has a larger angular size (190$^{\prime\prime}$) 
298: and is significantly older (1.8 $\times$ 10$^4$ yr). The reverse 
299: shock-heated ejecta material shows a relatively well-defined, but 
300: complex structure near the center of the SNR. The blast wave shock 
301: front is represented by the limb-brightened, soft shell-like feature 
302: surrounding the central ejecta. It appears that the reverse shock
303: has reached the center of the SNR in 0103$-$72.6, while it has
304: reached only $\sim$60\% of the current blast wave shock radius
305: in 0102.2$-$7219 \citep{gaetz00}. The blast wave shock front
306: of 0103$-$72.6 has swept up significant amounts of ambient ISM and 
307: slowed down ($v$~$\sim$~480~km~s$^{-1}$) to produce a 
308: limb-brightened soft X-ray shell, whereas 0102.2$-$7219 shows a 
309: high velocity ($v$~$\sim$~6000~km~s$^{-1}$; Hughes et al. 2000b)
310: shock front without significant limb-brigtening 
311: and presumably less swept-up ISM. Both of the SNRs have a remarkably 
312: circular outer boundary, and thus likely uniform surrounding ISM 
313: distributions. The morphological characteristics of these two SMC 
314: SNRs then appear to be primarily caused by the substantial difference 
315: in the ages ($\sim$10$^{3}$ yr vs. $\sim$10$^4$ yr): i.e., 
316: 0103$-$72.6 appears to be a more evolved version of 0102.2$-$7219.
317: 
318: \acknowledgments
319: This work has been supported in parts by NASA contract NAS8-01128. 
320: JPH was supported by {\it Chandra} grants GO2-3069X and GO3-4086X. 
321: POS was supported by NASA contract NAS8-39073.
322: 
323: \begin{thebibliography}{}
324: 
325: \bibitem[Anders \& Grevesse 1989]{anders89} Anders, E., \& Grevesse,
326: N. 1989, Geochimica et Cosmochimica Acta, 53, 197
327: 
328: \bibitem[Blair et al. 2000]{blair00} Blair, W. P., Morse, J. A.,
329: Raymond, J. C. et al. 2000, ApJ, 537, 667
330: 
331: \bibitem[Borkowski et al. 2001]{bor01} Borkowski, K. J., Lyerly, 
332: W. J., \& Reynolds, S. P. 2001, ApJ, 548, 820
333: 
334: \bibitem[Dopita \& Tuohy 1984]{dopita84} Dopita, M. A., \& Tuohy,
335: I. R. 1984, ApJ, 282, 135
336: 
337: \bibitem[Filipovi\'c et al. 1998]{fili98} Filipovi\'c, M. D.,
338: Pietsch, W., Haynes, R. F. et al. 1998, A\&AS, 127, 119
339: 
340: \bibitem[Gaetz et al. 2000]{gaetz00} Gaetz, T. J., Butt, Y. M.,
341: Edgar, R. et al. 2000, ApJ, 534, L47
342: 
343: \bibitem[Haberl et al. 2000]{haberl00} Haberl, F., Filipovi\'c, M. D.,
344: Pietsch, W., \& Kahabka, P. 2000, A\&AS, 142, 41
345: 
346: \bibitem[Hughes et al. 2000a]{hughes00a} Hughes, J. P., Rakowski, C. E.,
347: Burrows, D. N. et al. 2000a, ApJ, 528, L109
348: 
349: \bibitem[Hughes et al. 2000b]{hughes00b} Hughes, J. P., Rakowski, C. E.,
350: \& Decourchelle, A. 2000b, ApJ, 543, L61
351: 
352: \bibitem[Mathewson et al. 1983]{math83} Mathewson, D. S., Ford,
353: V. L., Dopita, M. A. et al. 1983, ApJS, 51, 345
354: 
355: \bibitem[Mills et al. 1982]{mills82} Mills, B. Y., Little, A. G.,
356: Durdin, J. M. et al. 1982, MNRAS, 200, 1007
357: 
358: \bibitem[Nomoto et al. 1997]{nomoto97} Nomoto, K., Hashimoto, M.,
359: Tsujimoto, T. et al. 1997, Nucl. Phys. A, 616, 79c
360: 
361: \bibitem[Park et al. 2002]{park02} Park, S., Roming, P. W. A., Hughes,
362: J. P. et al. 2002, ApJ, 564, L39
363: 
364: \bibitem[Park et al. 2003a]{park03a} Park, S., Hughes, J. P., Slane, P. O.
365: et al. 2003a, ApJL, Submitted
366: 
367: \bibitem[Park et al. 2003b]{park03b} Park, S., Hughes, J. P., Slane, P. O.
368: et al. 2003b, ApJ, 592, L41
369: 
370: \bibitem[Russell \& Dopita 1992]{russell92} Russell, S. C., \& Dopita,
371: M. A., 1992, ApJ, 384, 508
372: 
373: \bibitem[Sasaki et al. 2001]{sasaki01} Sasaki, M., Stadlbauer, F. X.,
374: Haberl, F. et al. 2001, A\&A, 365, L237
375: 
376: \bibitem[Seward \& Michell 1981]{seward81} Seward, F. D., \& Michell,
377: M. 1981, ApJ, 243, 736
378: 
379: \bibitem[Thielemann et al. 1996]{thielemann96} Thielemann, F.-K,
380: Nomoto, K., \& Hashimoto, M.-A. 1996, ApJ, 460, 408
381: 
382: \bibitem[Townsley et al. 2000]{townsley00} Townsley, L. K., Broos, P. S.,
383: Garmire, G. P. et al. 2000, ApJ, 534, L139
384: 
385: \bibitem[Townsley et al. 2002a]{townsley02a} Townsley, L. K., Broos, P. S.,
386: Nousek, J. A. et al. 2002a, Nuclear Instruments \& Methods in
387: Physics Research Section A, 486, 751
388: 
389: \bibitem[Townsley et al. 2002b]{townsley02b} Townsley, L. K., Broos, P. S.,
390: Chartas, G. et al. 2002b, Nuclear Instruments  \& Methods in Physics Research 
391: Section A, 486, 716
392: 
393: \bibitem[Yokogawa et al. 2002]{yoko02} Yokogawa, J., Imanish, K.,
394: Koyama, K. et al. 2002, PASJ, 54, 53
395: 
396: \end{thebibliography}
397: 
398: \begin{deluxetable}{lcc}
399: \footnotesize
400: \tablecaption{Spectral parameters of 0103$-$72.6.
401: \label{tbl:tab1}}
402: \tablewidth{0pt}
403: \tablehead{ & \colhead{ISM\tablenotemark{a}} & 
404: \colhead{Ejecta\tablenotemark{b}} }
405: \startdata
406: $N_H$ (10$^{21}$ cm$^{-2}$) & $<$ 1.2 & 1.1$^{+0.6}_{-0.4}$ \\
407: $kT$ (keV) & 0.30$^{+0.12}_{-0.05}$ & 0.53$^{+0.25}_{-0.18}$ \\
408: $n_et$ (10$^{11}$ cm$^{-3}$ s) & 1.6$^{+1.2}_{-0.7}$ & 1.4$^{+0.3}_{-0.3}$ \\
409: $EM$ (10$^{58}$ cm$^{-3}$) & 5.4$^{+1.1}_{-1.1}$ & 0.8$^{+0.3}_{-0.3}$ \\
410: $\chi$$^{2}$/$\nu$ & 169.0/137 & 86.7/58 \\
411: \enddata
412: \tablenotetext{a}{The best-fit plane-shock model from the combined four
413: shell regions.}
414: \tablenotetext{b}{The plane-shock component of the center region.}
415: \end{deluxetable}
416:  
417: \begin{deluxetable}{cccccccc}
418: \footnotesize
419: \tablecaption{Comparisons of the Abundance Ratios.
420: \label{tbl:tab2}}
421: \tablewidth{0pt}
422: \tablehead{ \colhead{Abundance} & \colhead{Center} & 
423: \colhead{SMC\tablenotemark{a}} & & & \colhead{Model\tablenotemark{b}} & & \\
424: \colhead{Ratio} & \colhead{0103$-$72.6} & & \colhead{15 M$_{\odot}$} & 
425: \colhead{18 M$_{\odot}$} & \colhead{20 M$_{\odot}$} & 
426: \colhead{25 M$_{\odot}$} & \colhead{40 M$_{\odot}$} }
427: \startdata
428: O/Ne & 4.1$^{+1.2}_{-1.1}$ & 5.8 & 13.0  & 5.3 & 7.1 & 5.9 & 16.0 \\ 
429: O/Mg & 31.5$^{+12.5}_{-14.7}$ & 11.2 & 15 & 24 & 12 & 19 & 26 \\
430: O/Si & 27.1$^{+12.8}_{-14.9}$ & 10.0 & 7.9 & 15 & 25 & 45 & 136 \\
431: \enddata
432: \tablenotetext{a}{Russell \& Dopita (1992).}
433: \tablenotetext{b}{Type II SN nucleosynthesis model by Nomoto et al. (1997).}
434: \end{deluxetable}
435: 
436: \begin{figure}[h]
437: \figurenum{1}
438: \centerline{\includegraphics[angle=0,width=3.5in]{f1.ps}}
439: \figcaption[]{ The true-color ACIS image of 0103$-$72.6. The red
440: is 0.3$-$0.75 keV, the green is 0.75$-$0.99 keV, and the blue
441: is 0.99$-$3.0 keV band images. Each broad-subband image has 
442: been adaptively smoothed. We note that the bright clumpy features
443: in the central region are real structures which were resolved
444: by the ACIS (see Figure~\ref{fig:fig3}). We however caution that faint, 
445: compact clumps ($\sim$arcsec scale) along the outer shell are likely
446: artifacts caused by the adaptive smoothing.
447: \label{fig:fig1}}
448: \end{figure}
449: 
450: \begin{figure}[h]
451: \figurenum{2}
452: \centerline{\includegraphics[angle=0,width=3.5in]{f2.ps}}
453: \figcaption[]{ The linear gray-scale O EW image of SNR 0103$-$72.6, 
454: comprised of combined O He$\alpha$ and O Ly$\alpha$ lines. 
455: The darker gray-scales indicate high EWs: i.e., $\sim$500$-$650 eV
456: around the central regions, and $\sim$150$-$300 eV otherwise. 
457: The line emission was extracted from $E$ = 500$-$720 eV. 
458: The underlying continuum was from $E$ = 300$-$450 eV and $E$ = 
459: 760$-$850 eV, and then logarithmically interpolated to the mean line
460: energy of 575 eV. Both line and continuum images have been adaptively
461: smoothed before the calculation of the EW. The EWs were set to zero
462: near the boundary of the SNR (as marked with a dashed line) in order 
463: to avoid the noise caused by low background photon statistics. 
464: \label{fig:fig2}}
465: \end{figure}
466: 
467: \begin{figure}[h]
468: \figurenum{3}
469: \centerline{\includegraphics[angle=0,width=3.5in]{f3.ps}}
470: \figcaption[]{ The gray-scale broadband image of SNR 0103$-$72.6. 
471: The image has been rebinned by 4 pixels for the purpose of display. 
472: Darker gray-scales are higher intensities. Regions used for the 
473: spectral analysis are marked.
474: \label{fig:fig3}}
475: \end{figure}
476: 
477: \begin{figure}[h]
478: \figurenum{4}
479: \centerline{\includegraphics[angle=0,width=0.5\textwidth]{f4.ps}}
480: \figcaption[]{X-ray spectrum from the swept-up ISM of 0103$-$72.6. 
481: The individual flux levels have been arbitrarily scaled for the
482: purpose of display. In the lower panel, residuals from the best-fit 
483: model for the east+west regions are presented with filled-circles, 
484: while those for the shell regions are presented as data points without 
485: filled-circles.
486: \label{fig:fig4}}
487: \end{figure}
488: 
489: \begin{figure}[h]
490: \figurenum{5}
491: \centerline{\includegraphics[angle=0,width=0.5\textwidth]{f5.ps}}
492: \figcaption[]{X-ray Spectrum of the O-rich ejecta as extracted from
493: the ``center'' region. The projected emission from the swept-up ISM
494: has been included with a Sedov component (the dotted line).
495: \label{fig:fig5}}
496: \end{figure}
497: 
498: \end{document}
499: