1: % Use only LaTeX2e, calling the article.cls class and 12-point type.

2: 

3: \documentclass[11pt,letterpaper]{scrartcl}

4: 

5: % Users of the {thebibliography} environment or BibTeX should use the

6: % scicite.sty package, downloadable from *Science* at

7: % www.sciencmag.org/misc/con-info.shtml .  This package should properly

8: % format in-text reference calls and reference-list numbers.

9: 

10: \usepackage{scicite}

11: \usepackage[varg]{txfonts}

12: \usepackage{graphicx}

13: 

14: % Use times if you have the font installed; otherwise, comment out the

15: % following line.

16: 

17: %\usepackage{times}

18: 

19: % The preamble here sets up a lot of new/revised commands and

20: % environments.  It's annoying, but please do *not* try to strip these

21: % out into a separate .sty file (which could lead to the loss of some

22: % information when we convert the file to other formats).  Instead, keep

23: % them in the preamble of your main LaTeX source file.

24: 

25: 

26: % The following parameters seem to provide a reasonable page setup.

27: 

28: \topmargin 0.0cm

29: \oddsidemargin 0.2cm

30: \textwidth 16cm 

31: \textheight 21cm

32: \footskip 1.0cm

33: 

34: 

35: %The next command sets up an environment for the abstract to your paper.

36: 

37: \newenvironment{sciabstract}{%

38: \begin{quote} \bf}

39: {\end{quote}}

40: 

41: 

42: % If your reference list includes text notes as well as references,

43: % include the following line; otherwise, comment it out.

44: 

45: \renewcommand\refname{References and Notes}

46: 

47: % The following lines set up an environment for the last note in the

48: % reference list, which commonly includes acknowledgments of funding,

49: % help, etc.  It's intended for users of BibTeX or the {thebibliography}

50: % environment.  Users who are hand-coding their references at the end

51: % using a list environment such as {enumerate} can simply add another

52: % item at the end, and it will be numbered automatically.

53: 

54: \newcounter{lastnote}

55: \newenvironment{scilastnote}{%

56: \setcounter{lastnote}{\value{enumiv}}%

57: \addtocounter{lastnote}{+1}%

58: \begin{list}%

59: {\arabic{lastnote}.}

60: {\setlength{\leftmargin}{.28in}}

61: {\setlength{\labelsep}{.5em}}}

62: {\end{list}}

63: 

64: 

65: % Include your paper's title here

66: 

67: \title{A Common Explosion Mechanism for Type Ia Supernovae} 

68: 

69: 

70: % Place the author information here.  Please hand-code the contact

71: % information and notecalls; do *not* use \footnote commands.  Let the

72: % author contact information appear immediately below the author names

73: % as shown.  We would also prefer that you don't change the type-size

74: % settings shown here.

75: 

76: \author

77: {Paolo~A.~Mazzali$^{1,2,3,4\ast}$,

78: Friedrich K.~R\"opke$^{1,5}$,\\

79: Stefano Benetti$^{6}$,

80: and Wolfgang Hillebrandt$^{1}$

81: \\

82: \\

83: \normalsize{$^{1}$Max-Planck Institut f\"ur Astrophysik,}\\

84: \normalsize{Karl-Schwarzschildstr.1, 85748 Garching, Germany}\\

85: \normalsize{$^{2}$Department of Astronomy, School of Science,}\\ 

86: \normalsize{University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan}\\

87: \normalsize{$^{3}$Research Center for the Early Universe, School of Science,}\\

88: \normalsize{University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan}\\

89: \normalsize{$^{4}$Istituto Nazionale di Astrofisica-OATs, Via Tiepolo 11,

90: I-34131 Trieste, Italy}\\ 

91: \normalsize{$^{5}$Department of Astronomy and Astrophysics, University

92: of California Santa Cruz,}\\

93: \normalsize{1156 High Street, Santa Cruz, CA 95064, U.S.A.}\\ 

94: \normalsize{$^{6}$Istituto Nazionale di Astrofisica-OAPd,}\\ 

95: \normalsize{vicolo dell'Osservatorio, 2, I-35122 Padova, Italy}\\

96: \\

97: \normalsize{$^\ast$To whom correspondence should be addressed; E-mail: mazzali@ts.astro.it}\\

98: % \normalsize{$^{\ast\ast}$To whom correspondence should be addressed;}

99: }

100: 

101: % Include the date command, but leave its argument blank.

102: 

103: \date{}

104: 

105: 

106: 

107: %%%%%%%%%%%%%%%%% END OF PREAMBLE %%%%%%%%%%%%%%%%

108: 

109: %%%%%%%%%%%%%%%% Authers' Defenision  %%%%%%%%%%%%%%%%%

110: 

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

112: \newcommand{\eg}{e.g.\ }

113: \newcommand{\ie}{i.e.\ }

114: \newcommand{\Msun}{M_{\odot}}

115: \newcommand{\Rsun}{R_{\odot}}

116: \newcommand{\Lsun}{L_{\odot}}

117: \newcommand{\kms}{km~s$^{-1}$}

118: \newcommand{\ergs}{erg~s$^{-1}$}

119: \newcommand{\Ha}{H$\alpha$}

120: \newcommand{\Hb}{H$\beta$}

121: \newcommand{\HI}{H~{\sc i}}

122: \newcommand{\HeI}{He~{\sc i}}

123: \newcommand{\OI}{O~{\sc i}}

124: \newcommand{\OII}{O~{\sc ii}}

125: \newcommand{\CII}{C~{\sc ii}}

126: \newcommand{\NaI}{Na~{\sc i}}

127: \newcommand{\MgII}{Mg~{\sc ii}}

128: \newcommand{\MgI}{Mg~{\sc i}}

129: \newcommand{\SII}{S~{\sc ii}}

130: \newcommand{\SiI}{Si~{\sc i}}

131: \newcommand{\SiII}{Si~{\sc ii}}

132: \newcommand{\SiIII}{Si~{\sc iii}}

133: \newcommand{\CaII}{Ca~{\sc ii}}

134: \newcommand{\TiII}{Ti~{\sc ii}}

135: \newcommand{\CrII}{Cr~{\sc ii}}

136: \newcommand{\FeI}{Fe~{\sc i}}

137: \newcommand{\FeII}{Fe~{\sc ii}}

138: \newcommand{\FeIII}{Fe~{\sc iii}}

139: \newcommand{\CoII}{Co~{\sc ii}}

140: \newcommand{\CoIII}{Co~{\sc iii}}

141: \newcommand{\NiII}{Ni~{\sc ii}}

142: \newcommand{\Feff}{$^{54}$Fe}

143: \newcommand{\Fefs}{$^{56}$Fe}

144: \newcommand{\Cofs}{$^{56}$Co}

145: \newcommand{\Nifs}{$^{56}$Ni}

146: \newcommand{\Nife}{$^{58}$Ni}

147: \newcommand{\Mej}{$M_{\rm ej}$}

148: \newcommand{\KE}{$E$}

149: \newcommand{\Dm}{$\Delta {\rm m_{15}(B)}$}

150: 

151: \def\gsim{\mathrel{\rlap{\lower 4pt \hbox{\hskip 1pt $\sim$}}\raise 1pt \hbox {$>$}}} 

152: \def\lsim{\mathrel{\rlap{\lower 4pt \hbox{\hskip 1pt $\sim$}}\raise 1pt \hbox {$<$}}}

153: 

154: \def\aj{AJ}%

155:           % Astronomical Journal

156: \def\araa{ARA\&A}%

157:           % Annual Review of Astron and Astrophys

158: \def\apj{ApJ}%

159:           % Astrophysical Journal

160: \def\apjl{ApJ}%

161:           % Astrophysical Journal, Letters

162: \def\apjs{ApJS}%

163:           % Astrophysical Journal, Supplement

164: \def\ao{Appl.~Opt.}%

165:           % Applied Optics

166: \def\apss{Ap\&SS}%

167:           % Astrophysics and Space Science

168: \def\aap{A\&A}%

169:           % Astronomy and Astrophysics

170: \def\aapr{A\&A~Rev.}%

171:           % Astronomy and Astrophysics Reviews

172: \def\aaps{A\&AS}%

173:           % Astronomy and Astrophysics, Supplement

174: \def\azh{AZh}%

175:           % Astronomicheskii Zhurnal

176: \def\baas{BAAS}%

177:           % Bulletin of the AAS

178: \def\jrasc{JRASC}%

179:           % Journal of the RAS of Canada

180: \def\memras{MmRAS}%

181:           % Memoirs of the RAS

182: \def\mnras{MNRAS}%

183:           % Monthly Notices of the RAS

184: \def\pra{Phys.~Rev.~A}%

185:           % Physical Review A: General Physics

186: \def\prb{Phys.~Rev.~B}%

187:           % Physical Review B: Solid State

188: \def\prc{Phys.~Rev.~C}%

189:           % Physical Review C

190: \def\prd{Phys.~Rev.~D}%

191:           % Physical Review D

192: \def\pre{Phys.~Rev.~E}%

193:           % Physical Review E

194: \def\prl{Phys.~Rev.~Lett.}%

195:           % Physical Review Letters

196: \def\pasp{PASP}%

197:           % Publications of the ASP

198: \def\pasj{PASJ}%

199:           % Publications of the ASJ

200: \def\qjras{QJRAS}%

201:           % Quarterly Journal of the RAS

202: \def\skytel{S\&T}%

203:           % Sky and Telescope

204: \def\solphys{Sol.~Phys.}%

205:           % Solar Physics

206: \def\sovast{Soviet~Ast.}%

207:           % Soviet Astronomy

208: \def\ssr{Space~Sci.~Rev.}%

209:           % Space Science Reviews

210: \def\zap{ZAp}%

211:           % Zeitschrift fuer Astrophysik

212: \def\nat{Nature}%

213:           % Nature

214: \def\iaucirc{IAU~Circ.}%

215:           % IAU Cirulars

216: \def\aplett{Astrophys.~Lett.}%

217:           % Astrophysics Letters

218: \def\apspr{Astrophys.~Space~Phys.~Res.}%

219:           % Astrophysics Space Physics Research

220: \def\bain{Bull.~Astron.~Inst.~Netherlands}%

221:           % Bulletin Astronomical Institute of the Netherlands

222: \def\fcp{Fund.~Cosmic~Phys.}%

223:           % Fundamental Cosmic Physics

224: \def\gca{Geochim.~Cosmochim.~Acta}%

225:           % Geochimica Cosmochimica Acta

226: \def\grl{Geophys.~Res.~Lett.}%

227:           % Geophysics Research Letters

228: \def\jcp{J.~Chem.~Phys.}%

229:           % Journal of Chemical Physics

230: \def\jgr{J.~Geophys.~Res.}%

231:           % Journal of Geophysics Research

232: \def\jqsrt{J.~Quant.~Spec.~Radiat.~Transf.}%

233:           % Journal of Quantitiative Spectroscopy and Radiative Trasfer

234: \def\memsai{Mem.~Soc.~Astron.~Italiana}%

235:           % Mem. Societa Astronomica Italiana

236: \def\nphysa{Nucl.~Phys.~A}%

237:           % Nuclear Physics A

238: \def\physrep{Phys.~Rep.}%

239:           % Physics Reports

240: \def\physscr{Phys.~Scr}%

241:           % Physica Scripta

242: \def\planss{Planet.~Space~Sci.}%

243:           % Planetary Space Science

244: \def\procspie{Proc.~SPIE}%

245:           % Proceedings of the SPIE

246: \let\astap=\aap

247: \let\apjlett=\apjl

248: \let\apjsupp=\apjs

249: \let\applopt=\ao

250: 

251: %%%%%%%%%%%%%%%%%

252: 

253: \begin{document} 

254: 

255: % Double-space the manuscript.

256: 

257: %\baselineskip24pt

258: 

259: % Make the title.

260: 

261: \maketitle 

262: 

263: 

264: 

265: % Place your abstract within the special {sciabstract} environment.

266: 

267: \begin{sciabstract}

268: Type Ia supernovae, the thermonuclear explosions of white dwarf stars

269: composed of carbon and oxygen, were instrumental as distance

270: indicators in establishing the acceleration of the universe's

271: expansion. However, the physics of the explosion are debated.  Here we

272: report a systematic spectral analysis of a large sample of well

273: observed type Ia supernovae. Mapping the velocity distribution of the

274: main products of nuclear burning, we constrain theoretical

275: scenarios. We find that all supernovae have low-velocity cores of

276: stable iron-group elements. Outside this core, nickel-56 dominates the

277: supernova ejecta. The outer extent of the iron-group material depends

278: on the amount of nickel-56 and coincides with the inner extent of

279: silicon, the principal product of incomplete burning. The outer extent

280: of the bulk of silicon is similar in all SNe, having an expansion

281: velocity of $\sim 11000\, \mathrm{km}\,\mathrm{s}^{-1}$ and corresponding

282: to a mass of slightly over one solar mass. This indicates that all the

283: supernovae considered here burned similar masses, and suggests that

284: their progenitors had the same mass. Synthetic light curve parameters

285: and three-dimensional explosion simulations support this

286: interpretation. A single explosion scenario, possibly a delayed

287: detonation, may thus explain most type Ia supernovae.

288: \end{sciabstract}

289: 

290: 

291: 

292: When a white dwarf (WD) composed of carbon and oxygen accreting mass

293: from a companion star in a binary system approaches the Chandrasekhar

294: mass [$M_{\mathrm{Ch}} \approx 1.38$ solar masses ($M_{\odot}$)], high

295: temperature causes the ignition of explosive nuclear burning reactions

296: that process stellar material and produce energy. The star explodes

297: leaving no remnant, producing a Type Ia supernova (SNIa)

298: \cite{nom84}. At high stellar material densities, burning yields

299: nuclear statistical equilibrium (NSE) isotopes, in particular

300: radioactive $^{56}$Ni which decays to $^{56}$Co and $^{56}$Fe making

301: the SN bright \cite{kuch94}. At lower densities intermediate mass

302: elements (IME) are synthesised. Both groups of elements are observed

303: in the optical spectra of SNeIa \cite{bra85}. An empirical relation

304: between an observed quantity, the B-magnitude decline over the first

305: 15 days after maximum luminosity [$\Delta m_{15} (B)$] and a physical

306: quantity, the SN maximum luminosity \cite{phil93}, can be used to

307: determine the distance to a SNIa. This method was applied to very

308: distant SNeIa leading to the discovery of the accelerating Universe

309: \cite{riess98,perl99}. How the explosion actually proceeds is however

310: debated \cite{reinecke02,gamezo04,plewa04}, as is the nature of the

311: progenitor system: Accretion may occur either from a more massive

312: companion (e.g. a giant) or via the merging of two carbon-oxygen WDs

313: \cite{hillnie00}. This casts a shadow on the reliability of SNeIa as

314: distance indicators, as intrinsically very different explosions may

315: result in the observed correlation.

316: 

317: We derive the distribution of the principal elements in 23 nearby

318: SNeIa (distances $<$$40 \, \mathrm{Mpc}$) with good spectral coverage

319: extending from before maximum to the late nebular phase, about one

320: year later. The sample (table S1) covers a wide range of light-curve

321: decline rates and includes peculiar objects like SN2000cx, which

322: violates the luminosity-decline-rate relation.

323: 

324: Because of the hydrodynamic properties of the explosion, the expansion

325: velocity of the ejecta is proportional to radius and serves as a

326: radial coordinate. As the SN expands, deeper layers are exposed.  The

327: outer layers, visible in the first few weeks after the explosion, are

328: dominated by IME. Because silicon is the most abundant IME, we measured its

329: characteristic expansion velocity from the blueshift of the absorption

330: core of the strong SiII $6355\,\mbox{\AA}$ line in all spectra where it was

331: visible.  This velocity decreases with time. Fitting the postmaximum

332: velocity evolution and extrapolating it to the earliest times, when

333: the outermost parts of the ejecta are visible, we derived the outer

334: extension of the bulk of Si. This represents a lower limit of the

335: outer extent of burning. The Si velocity

336: $v(\mathrm{Si})$ is similar in all SNeIa, regardless of their

337: luminosity: $v(\mathrm{Si}) = 11900 \pm 1300 \, \mathrm{km} \,

338: \mathrm{s}^{-1}$.

339: 

340: A few SNe, defined as High Velocity Gradient (HVG) SNe \cite{ben05}, are

341: responsible for most of the dispersion.  They have a rapidly

342: decreasing $v$(Si) before maximum and very high-velocity CaII lines,

343: possibly the result of high-velocity blobs that carry little mass and

344: kinetic energy but cause High-Velocity absorption Features (HVF,

345: \cite{maz05}) that can abnormally broaden the SiII line profile

346: \cite{maz05,tan06}. Although excluding them from the sample decreases the

347: dispersion in $v$(Si) significantly [$v(\mathrm{Si}) = 11300 \pm 650

348: \, \mathrm{km} \, \mathrm{s}^{-1}$], we include all these SNe in our

349: discussion.

350: 

351: The maximum Si velocity thus measured is a conservative

352: estimate. The deep absorption core is produced in layers of high

353: silicon abundance.  Silicon is present at higher velocities, indicated

354: by the wavelength of bluest absorption in the earliest spectrum of

355: each SN.  However, measuring the bluest absorption velocity yields a

356: large scatter because the earliest spectra have different epochs and

357: may be affected by HVFs \cite{maz05}. Our method reliably determines the outer

358: location of the bulk of IMEs. 

359: 

360: \begin{figure}[t]

361: \centerline{\includegraphics[width=0.65\linewidth]{Zorrodiag.eps}}

362: \caption{\footnotesize The Zorro Diagram: Distribution of the

363:   principal isotopic groups in SNeIa.  The enclosed mass (linked to

364:   velocity via the W7 explosion model) of different burning products

365:   is shown versus decline-rate parameter $\Delta m_{15} (B)$ (a proxy

366:   for SN luminosity).  Individual SNe are coloured according to their

367:   velocity evolution \cite{ben05}: HVG, blue; low

368:   velocity gradient (LVG), green; and faint, red.  Open circles indicate

369:   the mass of stable $^{54}$Fe+$^{58}$Ni for each SN; solid circles

370:   that of $^{56}$Ni, and open triangles the sum of these (total NSE

371:   mass). Crosses show the sum of NSE and IME mass, indicating the

372:   total mass burned. The IME mass is the difference between crosses

373:   and triangles.  $^{54}$Fe and $^{58}$Ni are found in roughly

374:   constant amounts in the deepest parts of all SNe, irrespective of

375:   luminosity: $M (\mathrm{stable NSE}) = 0.20 \pm 0.05 \, M_{\odot}$

376:   (lower horizontal shaded area).  The $^{56}$Ni mass determines the

377:   SN luminosity. It correlates with $\Delta m_{15} (B)$:

378:   $M(^{56}\mathrm{Ni}) = 1.34 - 0.67 \Delta m_{15} (B) [M_{\odot}]$,

379:   with rms dispersion $0.13 \, M_{\odot}$ (lower diagonal shaded

380:   area).  The total NSE mass correlates with $\Delta m_{15} (B)$

381:   better than M($^{56}$Ni): $M(\mathrm{NSE}) = 1.55 - 0.69 \Delta

382:   m_{15} (B) [M_{\odot}]$, rms dispersion $0.09\, M_{\odot}$ (upper

383:   diagonal shaded area).  IME lie mostly outside the iron-group zone.

384:   The outer Si velocity is similar for all SNe except HVG SNe.  The

385:   mass enclosed by IMEs represents the total burned mass. When all SNe

386:   are included the average value is $\sim$$1.05 \pm 0.09 \, M_{\odot}$

387:   (upper horizontal shaded area).  Excluding HVG SNe the value is

388:   $\sim$$1.01 \pm 0.05 \, M_{\odot}$ (horizontal line).  Both values

389:   are independent of $\Delta m_{15} (B)$.  For a version of this plot

390:   with SN names and a bar diagram, see figs. S2 and S3.

391: \label{fig:Zorro}}

392: \end{figure}

393: 

394: The inner extent of Si, determined from the asymptotic velocity

395: of the SiII 6355$\, \mbox{\AA}$ line in post-maximum spectra

396: [fig.~1 in \cite{ben05}], is a steep function of $\Delta

397: m_{15} (B)$. The brightest (slowest declining) SNe have the thinnest

398: Si zones.

399: 

400: The inner ejecta, dominated by NSE elements, are best observed $\sim$1

401: year after the explosion, when dilution caused by expansion makes the

402: SN behave like a nebula, exposing the deepest layers.  Collisions with

403: the fast particles produced by the decay

404: $^{56}$Ni$\,\to\,$$^{56}$Co$\,\to\,$$^{56}$Fe heat the gas, which

405: cools emitting radiation mostly in forbidden lines.

406: 

407: We modeled the nebular spectra using a code that computes line

408: emission balancing heating and cooling in non-local thermodynamic

409: equilibrium (NLTE) \cite{maz01a}, including density and abundance

410: stratification. We adopted the density-velocity distribution of the

411: standard, one-dimensional, $M_\mathrm{Ch}$ explosion model W7

412: \cite{nom84}, (fig. S1). In the nebular phase the gas is transparent,

413: and line emissivity depends on the mass of the emitting

414: ion. Accurately estimating this mass requires determining the

415: ionization state of the gas. Forbidden lines of FeII and FeIII

416: dominate SNIa nebular spectra, reflecting the high abundance of NSE

417: material. Fe is mostly the product of $^{56}$Ni decay, which provides

418: heating. The stable, neutron-rich isotopes $^{54}$Fe and $^{58}$Ni do

419: not contribute to heating, but do contribute to cooling, because they

420: also emit forbidden lines. Their presence affects the ionization

421: balance. Both are mostly produced deep in the WD, at the highest

422: densities. The $^{54}$Fe nebular lines have wavelengths

423: indistinguishable from those of $^{56}$Fe.  

424: 

425: We determined the mass and

426: distribution in velocity of the Fe isotopes (and thus of $^{56}$Ni)

427: simultaneously fitting the ratio of the two strongest Fe

428: emissions(fig.~S1). One, near 4700$\, \mbox{\AA}$ includes both FeII and

429: FeIII lines, while the other, near 5200$\, \mbox{\AA}$, is only due to

430: FeII. A low upper limit to the mass of $^{58}$Ni is set by the

431: absence of strong emission lines (in particular at 7380$\, \mbox{\AA}$).

432: 

433: In the deepest layers (Fig.~\ref{fig:Zorro}), all SNe contain

434: $\sim$0.1 to $0.3 \, M_{\odot}$ of stable NSE isotopes, with a large

435: scatter and no dependence on $\Delta m_{15} (B)$ [see also

436: \cite{woo06}]. 

437: 

438: As expected, the $^{56}$Ni mass correlates inversely with $\Delta

439: m_{15} (B)$, ranging from $0.9 \, M_{\odot}$ for the slowest declining

440: (most luminous) SNe to $0.1\, M_{\odot}$ for the fastest declining

441: (dimmest) ones. The root mean square (rms) dispersion is $0.13

442: \,M_{\odot}$, but SNe with intermediate decline rates [$\Delta m_{15}

443: (B) \sim 1.05$ to $1.5~\mathrm{magnitudes}$] show variations of almost a factor of 2

444: for the same value of $\Delta m_{15} (B)$.  These SNe could cause

445: scatter about the mean luminosity-decline-rate relation \cite{maz06}.

446: 

447: Once the contributions of $^{56}$Ni, $^{54}$Fe, and

448: $^{58}$Ni are added together to evaluate the total NSE mass, the

449: dispersion decreases to $0.09\, M_{\odot}$. If SNe with different

450: amounts of $^{56}$Ni, and thus presumably different temperatures, but

451: similar NSE content have similar $\Delta m_{15} (B)$, it is likely

452: that abundances \cite{maz01} rather than temperature \cite{kasen06}

453: primarily determine the opacity and light-curve shape.

454: 

455: The outer velocity of the NSE region, determined from the width of the

456: Fe lines, correlates with SN luminosity \cite{maz98}. It coincides with the

457: innermost silicon velocity, marking the transition from complete to

458: incomplete burning: the difference between these two velocities,

459: $\Delta(v) = 650 \pm 900 \, \mathrm{km} \, \mathrm{s}^{-1}$, is

460: consistent with zero. Remarkably, this results from two different

461: methods applied to data obtained almost 1 year apart. 

462: 

463: Thus, while the mass of $^{56}$Ni, and consequently the SN luminosity, can

464: differ significantly, other characteristics of SNeIa are remarkably

465: homogeneous.  In particular, the narrow dispersion of the outer Si

466: velocity indicates a similar extent of thermonuclear burning in all

467: SNeIa: SNe that produce less $^{56}$Ni synthesize more IMEs. 

468: 

469: The simplest interpretation of these seemingly antithetic results is

470: that thermonuclear burning consumes similar masses in all SNeIa.  We

471: explore whether this scenario is consistent with the $M_{\mathrm{Ch}}$

472: model \cite{hillnie00}.  Applying the density-velocity structure of

473: model W7 \cite{nom84}, we transform $v$(Si) to mass

474: (Fig.~\ref{fig:Zorro}). We find that the outer shell of silicon

475: encloses a mass of at least $\sim 1.05 \pm 0.09 \, M_{\odot}$ ($\sim

476: 1.01 \pm 0.05 \, M_{\odot}$, excluding HVG SNe), independently of

477: $\Delta m_{15} (B)$. This is a lower limit to the burned mass.

478: 

479: 

480: The light-curve width $\tau$ is related to $\Delta m_{15} (B)$

481: \cite{gold01} and depends on the ejected mass $M_{\mathrm{ej}}$, the

482: kinetic energy $E_{\mathrm{k}}$, and the opacity $\kappa$ as $\tau

483: \propto \kappa^{1/2} E_{\mathrm{k}}^{-1/4} M_{\mathrm{ej}}^{3/4}$

484: \cite{arn82}. We test our assumption computing parametrized

485: light-curve widths and comparing them to observed values. The

486: resulting luminosity-decline-rate relation (Fig.~\ref{fig:LCprops}) is

487: very tight, and practically identical to the observed one.  This

488: result supprots our opacity parametrization, corroborating our

489: hypothesis that the mass burned is similar in all SNeIa. Given its

490: weak dependence on $E_{\mathrm{k}}$, the light-curve width is not much

491: affected if more of the outer part of the WD is burned to IME, as

492: these do not contribute much to the opacity.

493: 

494: If SNeIa burn a similar mass, the progenitor mass is also likely to

495: be the same, namely $M_{\mathrm{Ch}}$. Because the outcome of the

496: burning depends essentially on fuel density, a variation in iron-group

497: elements production in $M_{\mathrm{Ch}}$ WDs requires different WD

498: expansion histories.  This in turn depends on the details of the

499: burning. Once the WD reaches $M_{\mathrm{Ch}}$, a thermonuclear flame

500: is ignited near the center.  The flame must start as a subsonic

501: deflagration \cite{nom84}, mediated by microphysical transport and

502: accelerated by turbulence. As it propagates outwards, it could undergo

503: a deflagration-to-detonation transition (DDT) and continue as a

504: shock-driven supersonic detonation wave, in a so-called delayed

505: detonation \cite{khokhlov91}. This constitutes the most extreme

506: explosion scenario admissible, exploring the limits of

507: $M_{\mathrm{Ch}}$ explosions.

508: 

509: \begin{table}[t]

510: \caption{Results and DDT parameters of parametrized three-dimensional delayed detonation

511:   models. \label{tab:res}}

512: \centerline{\begin{tabular}{llllll}

513: \hline

514: model  & $E_{\mathrm{kin}}^{\mathrm{asympt}} \, (\mathrm{10}^{51} erg)$ & $M(\mathrm{NSE}) \,

515: (M_\odot$) & $M(\mathrm{IME}) \, (M_\odot)$ & $t_{\mathrm{DDT}}$ (s) &

516: $\rho_{\mathrm{DDT}}$ ($10^7 \, \mathrm{g} \, \mathrm{cm}^{-3}$)\\

517: \hline

518: \emph{D800} & 1.004 & 0.638 & 0.547 & 0.675 & 2.40\\

519: \emph{D20}  & 1.237 & 0.833 & 0.435 & 0.724 & 1.92\\

520: \emph{D5}   & 1.524 & 1.141 & 0.220 & 0.731 & 1.33\\

521: \hline

522: \end{tabular}}

523: \end{table}

524: 

525: We modeled the explosion using a three-dimensional level-set approach

526: \cite{golombek05}. 

527: The ignition of the deflagration flame was treated as a stochastic

528: process generating a number of ignition spots placed randomly and

529: isotropically within $180\, \mathrm{km}$ of the WD centre \cite{woos}. 

530: 

531: What would cause the DDT is unclear. We assumed that it occurs as

532: turbulence penetrates the internal flame structure: the onset of the

533: so-called distributed burning regime \cite{niewoo97}. This happens at low fuel

534: densities, after some WD pre-expansion in the deflagration phase. The

535: detonation is triggered artificially where the chosen DDT criterion is

536: first satisfied, typically near the outer edges of the deflagration

537: structure (Table~\ref{tab:res} shows DDT parameters). 

538: Three simulations, with 800, 20, and 5 ignition spots, termed D800,

539: D20, and D5, respectively, were performed on a moving cellular

540: Cartesian grid \cite{roep06} comprising the full star. Model D800, with its

541: dense distribution of ignition points, exhausts the carbon-oxygen fuel

542: at the WD centre almost completely in the deflagration phase. The

543: energy release quickly expands the star, and the subsequent detonation

544: mainly transforms low density outer material to IME. In contrast, the

545: few ignition spots of model D5 consume little material during the

546: deflagration, leaving more fuel at high densities which is converted

547: mostly to NSE isotopes in the vigorous detonation phase. Model D20 provides an

548: intermediate case. 

549: 

550: The ejecta compositions of the model explosions agree grossly with the

551: results derived from the spectra. The NSE mass produced ranges from

552: 0.638 to 1.141$\, M_{\odot}$.  The weakest explosions result from

553: optimal burning in the deflagration phase \cite{roep06}, and

554: subluminous SNeIa are not reached in our parametrization. However,

555: the conditions for DDT need further investigation. Eventually these

556: events may be explained within a single framework. Some extremely

557: luminous SNeIa may come from very rapidly rotating WDs whose mass

558: exceeds \cite{how06}, but these are rare.

559: 

560: The distribution derived from observations of burning products inside

561: SNeIa could result from the variation of a single initial parameter,

562: the flame ignition configuration, in M$_{\mathrm{Ch}}$ delayed detonations.  The

563: luminosity-decline-rate relation can be reproduced using this

564: distribution and a simple opacity parametrization. Our results support

565: the M$_{\mathrm{Ch}}$ scenario for most SNeIa, adding confidence to their use as

566: distance indicators.

567: 

568: \begin{figure}[t]

569: \centerline{\includegraphics[width=0.75\linewidth]{LCprops.eps}}

570: \caption{\footnotesize Observed and synthetic luminosity-decline-rate

571: relations for the SNe in our sample. Colors indicate velocity

572: evolution as in Fig.~\ref{fig:Zorro}.  The peak luminosity $L$ was computed from

573: M($^{56}$Ni) as $L = 2 \times 10^{43} \, M(^{56}\mathrm{Ni})$

574: \cite{stritz}.  Observed bolometric light-curve (LC) widths were

575: obtained from observed rise and decline times \cite{contardo00} as $T

576: = (t_{-1/2}+ t_{+1/2})$.  When $t_{-1/2}$ is missing, it was estimated

577: from the relation between $t_{-1/2}$ and $\Delta m_{15} (B)$ derived

578: from the other SNe.  The dashed line is a linear regression between

579: luminosity and observed LC width for the seven SNe common to our sample

580: and that of \cite{contardo00}.  Synthetic bolometric LC widths were

581: obtained assuming that: (i) The LC width $\tau$ depends on ejected mass

582: $M_{\mathrm{ej}}$, kinetic energy $E_{\mathrm{k}}$, and opacity

583: $\kappa$ as: $\tau \propto \kappa^{1/2} E_{\mathrm{k}}^{-1/4}

584: M_{\mathrm{ej}}^{3/4}$ \cite{arn82}. (ii) For all SNe, $M_{\mathrm{ej}}

585: = M_{\mathrm{Ch}}$. The W7 \cite{nom84} density-velocity distribution

586: was used. (iii) The explosion kinetic energy depends on the burning

587: product: $ E_{\mathrm{k}} = [1.56 M(^{56}\mathrm{Ni}) + 1.74

588: M(\mathrm{stable NSE}) + 1.24 M(\mathrm{IME}) - 0.46] \times

589: 10^{51}\,\mathrm{erg}$ \cite{maz01}. (iv) Opacity is mostly due to line

590: absorption \cite{paul97}. Accordingly, NSE elements contribute much

591: more than IMEs because their atomic level structure is more complex. The

592: opacity was therefore parametrised according to the abundances of

593: different species: $\kappa \propto M(\mathrm{NSE}) + 0.1\,

594: M(\mathrm{IME})$ \cite{maz06}.  To compare the parametrized LC widths

595: $\tau$ to observed values T, a scale factor $x=T/\tau$ was computed

596: for each of the seven SNe common to our sample and that of

597: \cite{contardo00}. The average factor, $x=24.447$, was used to scale

598: all SNe. Dots show the individual SNe. The continuous line is the

599: linear regression between luminosity and synthetic LC width for our 23

600: SNeIa sample.  For a version of this plot with SN names, see fig.~S4.

601: \label{fig:LCprops}}

602: \end{figure}

603: 

604:  

605: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

606: 

607: \begin{thebibliography}{10}

608: 

609: \bibitem{nom84}

610: K.~{Nomoto}, F.-K. {Thielemann}, K.~{Yokoi}, {\it \apj\/} {\bf 286}, 644

611:   (1984).

612: 

613: \bibitem{kuch94}

614: M.~J. {Kuchner}, R.~P. {Kirshner}, P.~A. {Pinto}, B.~{Leibundgut}, {\it

615:   \apjl\/} {\bf 426}, L89 (1994).

616: 

617: \bibitem{bra85}

618: D.~{Branch}, J.~B. {Doggett}, K.~{Nomoto}, F.-K. {Thielemann}, {\it \apj\/}

619:   {\bf 294}, 619 (1985).

620: 

621: \bibitem{phil93}

622: M.~M. {Phillips}, {\it \apjl\/} {\bf 413}, L105 (1993).

623: 

624: \bibitem{riess98}

625: A.~G. {Riess}, {\it et~al.\/}, {\it \aj\/} {\bf 116}, 1009 (1998).

626: 

627: \bibitem{perl99}

628: S.~{Perlmutter}, {\it et~al.\/}, {\it \apj\/} {\bf 517}, 565 (1999).

629: 

630: \bibitem{reinecke02}

631: M.~{Reinecke}, W.~{Hillebrandt}, J.~C. {Niemeyer}, {\it \aap\/} {\bf 391}, 1167

632:   (2002).

633: 

634: \bibitem{gamezo04}

635: V.~N. {Gamezo}, A.~M. {Khokhlov}, E.~S. {Oran}, {\it Physical Review Letters\/}

636:   {\bf 92}, 211102 (2004).

637: 

638: \bibitem{plewa04}

639: T.~{Plewa}, A.~C. {Calder}, D.~Q. {Lamb}, {\it \apjl\/} {\bf 612}, L37 (2004).

640: 

641: \bibitem{hillnie00}

642: W.~{Hillebrandt}, J.~C. {Niemeyer}, {\it \araa\/} {\bf 38}, 191 (2000).

643: 

644: \bibitem{ben05}

645: S.~{Benetti}, {\it et~al.\/}, {\it \apj\/} {\bf 623}, 1011 (2005).

646: 

647: \bibitem{maz05}

648: P.~A. {Mazzali}, {\it et~al.\/}, {\it \apjl\/} {\bf 623}, L37 (2005).

649: 

650: \bibitem{tan06}

651: M.~{Tanaka}, P.~A. {Mazzali}, K.~{Maeda}, K.~{Nomoto}, {\it \apj\/} {\bf 645},

652:   470 (2006).

653: 

654: \bibitem{maz01a}

655: P.~A. {Mazzali}, K.~{Nomoto}, F.~{Patat}, K.~{Maeda}, {\it \apj\/} {\bf 559},

656:   1047 (2001).

657: 

658: \bibitem{woo06}

659: S.~E. {Woosley}, D.~{Kasen}, S.~{Blinnikov}, E.~{Sorokina}, {\it ArXiv

660:   Astrophysics e-prints astro-ph/0609562\/}  (2006).

661: 

662: \bibitem{maz06}

663: P.~A. {Mazzali}, P.~{Podsiadlowski}, {\it \mnras\/} {\bf 369}, L19 (2006).

664: 

665: \bibitem{maz01}

666: P.~A. {Mazzali}, {\it et~al.\/}, {\it \apj\/} {\bf 547}, 988 (2001).

667: 

668: \bibitem{kasen06}

669: D.~{Kasen}, S.~E. {Woosley}, {\it astro-ph/0609540\/}  (2006).

670: 

671: \bibitem{maz98}

672: P.~A. {Mazzali}, E.~{Cappellaro}, I.~J. {Danziger}, M.~{Turatto}, S.~{Benetti},

673:   {\it \apjl\/} {\bf 499}, L49 (1998).

674: 

675: \bibitem{gold01}

676: G.~{Goldhaber}, {\it et~al.\/}, {\it \apj\/} {\bf 558}, 359 (2001).

677: 

678: \bibitem{arn82}

679: W.~D. {Arnett}, {\it \apj\/} {\bf 253}, 785 (1982).

680: 

681: \bibitem{khokhlov91}

682: A.~M. {Khokhlov}, {\it \aap\/} {\bf 245}, 114 (1991).

683: 

684: \bibitem{golombek05}

685: I.~{Golombek}, J.~C. {Niemeyer}, {\it \aap\/} {\bf 438}, 611 (2005).

686: 

687: \bibitem{woos}

688: S.~E. {Woosley}, S.~{Wunsch}, M.~{Kuhlen}, {\it \apj\/} {\bf 607}, 921 (2004).

689: 

690: \bibitem{niewoo97}

691: J.~C. {Niemeyer}, S.~E. {Woosley}, {\it \apj\/} {\bf 475}, 740 (1997).

692: 

693: \bibitem{roep06}

694: F.~K. {R{\"o}pke}, W.~{Hillebrandt}, J.~C. {Niemeyer}, S.~E. {Woosley}, {\it

695:   \aap\/} {\bf 448}, 1 (2006).

696: 

697: \bibitem{how06}

698: D.~A. {Howell}, {\it et~al.\/}, {\it \nat\/} {\bf 443}, 308 (2006).

699: 

700: \bibitem{stritz}

701: M.~{Stritzinger}, B.~{Leibundgut}, {\it \aap\/} {\bf 431}, 423 (2005).

702: 

703: \bibitem{contardo00}

704: G.~{Contardo}, B.~{Leibundgut}, W.~D. {Vacca}, {\it \aap\/} {\bf 359}, 876

705:   (2000).

706: 

707: \bibitem{paul97}

708: A.~W.~A. {Pauldrach}, {\it et~al.\/}, {\it \aap\/} {\bf 312}, 525 (1996).

709: 

710: \end{thebibliography}

711: 

712: % Following is a new environment, {scilastnote}, that's defined in the

713: % preamble and that allows authors to add a reference at the end of the

714: % list that's not signaled in the text; such references are used in

715: % *Science* for acknowledgments of funding, help, etc.

716: 

717: \begin{scilastnote}

718: \item We thank Elena Pian and Daniel Sauer for help with data analysis.\\

719: This work was partly supported by the European Union's Human Potential

720: Programme under contract 

721: HPRN-CT-2002-00303, ``The Physics of Type Ia Supernovae''. 

722: \end{scilastnote}

723: 

724: 

725: 

726: % For your review copy (i.e., the file you initially send in for

727: % evaluation), you can use the {figure} environment and the

728: % \includegraphics command to stream your figures into the text, placing

729: % all figures at the end.  For the final, revised manuscript for

730: % acceptance and production, however, PostScript or other graphics

731: % should not be streamed into your compliled file.  Instead, set

732: % captions as simple paragraphs (with a \noindent tag), setting them

733: % off from the rest of the text with a \clearpage as shown  below, and

734: % submit figures as separate files according to the Art Department's

735: % instructions.

736: 

737: 

738: 

739: \end{document}

740: 

741: 

742: 

743: 

744: 

745: 

746: 

747: 

748: 

749: 

750: 

751: 

752: 

753: 

754: 

755: 

756: 

757: 

758: 

759: 

760: