1: % aa.dem
2: % AA vers. 6.1, LaTeX class for Astronomy & Astrophysics
3: % demonstration file
4: % (c) Springer-Verlag HD
5: % revised by EDP Sciences
6: %-----------------------------------------------------------------------
7: %
8: %\documentclass[referee]{aa} % for a referee version
9: %\documentclass[onecolumn]{aa} % for a paper on 1 column
10: %\documentclass[longauth]{aa} % for the long lists of affiliations
11: %\documentclass[rnote]{aa} % for the research notes
12: %\documentclass[letter]{aa} % for the letters
13: %
14: \documentclass{aa}
15: %
16: \usepackage{graphicx}
17: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18: \usepackage{txfonts}
19: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20: %
21: \begin{document}
22: %
23: \title{Upper limit for circumstellar gas around the Type Ia
24: SN~2000cx}
25:
26: \author{F. Patat
27: \inst{1},
28: S. Benetti\inst{2},
29: S. Justham\inst{3},
30: P.A. Mazzali\inst{4,5},
31: L. Pasquini\inst{1},
32: E. Cappellaro\inst{2},\\
33: M. Della Valle\inst{6},
34: Ph.\ Podsiadlowski\inst{3},
35: M. Turatto\inst{2},
36: A. Gal-Yam\inst{7}
37: \and
38: J.D. Simon\inst{7}\fnmsep\thanks{Based on observations obtained
39: at ESO-Paranal.}
40: }
41:
42: \offprints{F. Patat}
43:
44: \institute{European Organization for Astronomical Research in the
45: Southern Hemisphere; K. Schwarzschild-Str. 2, 85748 Garching, Germany\\
46: \email{fpatat@eso.org}
47: \and
48: INAF - Osservatorio Astronomico di Padova, v. Osservatorio n.5,
49: 35122 Padova, Italy
50: \and
51: Department of Astrophysics, University of Oxford, Oxford OX1 3RH, UK
52: \and
53: Max-Planck-Institut fuer Astrophysik, K. Schwarzschild Str.2,
54: 85748, Garching, Germany
55: \and
56: INAF - Osservatorio Astronomico di Trieste, v. Tiepolo 11,
57: 34131 Trieste, Italy
58: \and
59: INAF - Osservatorio Astronomico di Arcetri, Largo E. Fermi n.5,
60: Firenze, Italy
61: \and
62: Astronomy Department, MS 105-24, California Institute of
63: Technology, Pasadena, CA 91125,USA\\
64: }
65:
66: \date{Received August 1, 2007; accepted August 27, 2007}
67:
68: % \abstract{}{}{}{}{}
69: % 5 {} token are mandatory
70:
71: \abstract
72: % context heading (optional)
73: % {} leave it empty if necessary
74: {The nature of the companion stars in Type Ia Supernova (SNe) progenitor
75: systems remains unclear. One possible way to discriminate between different
76: scenarios is the presence (or absence) of circumstellar material, the
77: left overs from the progenitor evolution that may be revealed by their
78: interaction with the SN.}
79: % aims heading (mandatory)
80: {A new method to probe the circumstellar environment
81: has been exploited for the normal Type Ia SN~2006X, leading for the first
82: time to the direct detection of material which escaped the progenitor
83: system. In this paper we apply the same analysis to the peculiar
84: Type Ia SN~2000cx, with the aim of constraining the properties of its
85: progenitor system.}
86: % methods heading (mandatory)
87: {Using multi-epoch, high-resolution spectroscopy we have studied the
88: spectral region where narrow, time-variable Na~ID absorption features are
89: expected in case circumstellar material is present along the line of
90: sight.}
91: % results heading (mandatory)
92: {No Na~ID absorption is detected in the rest-frame of the host galaxy
93: to a level of a few m\AA, setting a stringent upper limit to the column
94: density of the absorbing material
95: ($N(NaI)\leq $2$\times$10$^{10}$ cm$^{-2}$).}
96: % conclusions heading (optional), leave it empty if necessary
97: {In this respect the peculiar Type Ia SN~2000cx is different
98: from the normal Ia SN~2006X. Whether this
99: is to be attributed to a different progenitor system, to viewing-angle
100: effects or to a low metallicity remains to be clarified.}
101:
102: \keywords{supernovae: general - supernovae: individual (SN2000cx)
103: - interstellar medium - galaxies: individual (NGC524)
104: }
105:
106: \authorrunning{F. Patat et al.}
107:
108: \maketitle
109: %
110: %________________________________________________________________
111:
112: \section{Introduction}
113: \label{sec:intro}
114:
115: Due to their enormous luminosities and their homogeneity, Type Ia
116: Supernovae (hereafter SNe Ia) have been used in cosmology as standard
117: candles, with the ambitious aim of tracing the evolution of the
118: Universe (Riess et al. \cite{riess}; Perlmutter et
119: al. \cite{perlmutter}). Despite of the progresses made in this field,
120: the nature of the progenitor stars and the physics which governs these
121: powerful explosions are still uncertain. In general, they are thought
122: to originate from a close binary system (Whelan \& Iben
123: \cite{whelan}), where a white dwarf accretes material from a companion
124: until it approaches the Chandrasekhar limit and finally undergoes a
125: thermonuclear explosion. This scenario is widely accepted, but the
126: nature of both the accreting and the donor star is not yet known, even
127: though favorite configurations do exist (Branch et al. \cite{branch};
128: Parthasarathy et al. \cite{partha}; Tutukov \& Fedorova
129: \cite{tutukov}). A discriminant between some of the proposed scenarios
130: would be the detection of circumstellar material (CSM). However,
131: notwithstanding the importance of the quest, all attempts of detecting
132: direct signatures of the material being transferred to the accreting
133: white dwarf in normal SNe~Ia were so far frustrated, and only upper
134: limits to the mass-loss rate could be placed from optical (Mattila et
135: al.\ \cite{mattila}), radio (Panagia et al.\ \cite{panagia06}) and
136: UV/X-Ray emission (Immler et al. \cite{immler}). Claims of possible
137: ejecta-CSM interaction have been made for a few normal objects, namely
138: SN~1999ee (Mazzali et al. \cite{mazzali05}), SN~2001el (Wang et al.
139: \cite{wang03}), SN~2003du (Gerardy et al.\ \cite{gerardy}) and
140: SN~2005cg (Quimby et al. \cite{quimby06}). In all those cases, the
141: presence of CSM is inferred from the detection of high-velocity
142: components in the SN spectra. However, it must be noticed that these
143: features can be explained by a 3D structure of the explosion (Mazzali
144: et al. \cite{mazzali05}) and, therefore, circumstellar interaction is
145: not necessarily a unique interpretation (Quimby et
146: al. \cite{quimby06}). Furthermore, even if the high-velocity
147: components were indeed caused by an ejecta-CSM interaction, it would
148: not be possible to estimate the velocity or density of the CSM.
149:
150:
151: Two remarkable exceptions are represented by the peculiar SN~2002ic
152: and SN~2005gj, which have shown extremely pronounced hydrogen emission
153: lines (Hamuy et al. \cite{hamuy}; Aldering et al. \cite{aldering};
154: Prieto et al. \cite{prieto}), that have been interpreted as a sign of
155: strong ejecta-CSM interaction. However, the classification of these
156: supernovae as SNe Ia has been questioned (Benetti et
157: al. \cite{benetti06}), and even if they were SN Ia, they must be rare
158: and hence unlikely to account for normal Type Ia explosions (Panagia
159: et al. \cite{panagia06}). As a matter of fact, the only genuine
160: detection may be represented by the underluminous SN~2005ke (Patat et
161: al. \cite{patat05}), which has shown unprecedented X-ray emission
162: at the 3.6$\sigma$-level accompanied by a large UV excess (Immler et
163: al. \cite{immler}). These observations have been interpreted as the signature
164: of a possible weak interaction between the SN ejecta and material lost
165: by a companion star (Immler et al. \cite{immler}). Interestingly, the
166: SN has not been detected at radio wavelengths (Soderberg
167: \cite{soderberg}).
168:
169:
170: A different approach has been recently proposed and tested on the
171: normal SN Ia 2006X (Patat et al. \cite{patat}; hereafter P07). For
172: SN~2006X, P07 detected material within a few 10$^{16}$ cm from the
173: explosion site. Based on the velocity, density and location of the
174: CSM, P07 have concluded that the favorite companion star for the
175: progenitor of SN~2006X is a red giant at the time of explosion. The
176: method is based on the study of time variation of absorption lines
177: (Na~I, Ca~II, K~I) arising in the CSM. Since in SNe Ia the UV
178: bluewards of 3500\AA\/ undergoes severe line blocking by heavy
179: elements like Fe, Co, Ti and Cr (Pauldrach et al. \cite{pauldrach};
180: Mazzali \cite{mazzali00}), they are able to ionize any existing
181: circumstellar material only within a rather small radius. Once the SN
182: UV flux has significantly decreased in the post-maximum phase, the
183: material that has a sufficiently high density recombines, producing
184: time-variable absorption features. In particular, due to their
185: strength, the high-resolution study of the evolution of Na~I D lines
186: offers a powerful diagnostic, capable of revealing very small amounts
187: of material along the line of sight, without requiring direct
188: interaction between the SN ejecta and CSM.
189:
190: In order to investigate the existence of multiple channels to Type Ia
191: explosions and the possible presence of viewing angle effects (which
192: are expected if the material is confined within a disk or a torus), a
193: large sample of Type Ia events needs to be studied. The drawback of
194: the novel approach proposed by P07 is that it requires multi-epoch,
195: high-resolution ($\Delta\lambda/\lambda \geq$40,000) and high
196: signal-to-noise ratio ($SNR\geq$50) data, making it feasible only with
197: a large amount of integration time at large telescopes and for
198: close-by objects only. This is also one of the reasons that make the
199: data set of SN~2006X unique, at least in this respect.
200:
201: A few Type Ia SNe have been observed at high spectral resolution with
202: the aim of detecting possible signs of interaction with CSM (Lundqvist
203: et al. \cite{lundqvist03, lundqvist05}; Sollerman et
204: al. \cite{sollerman05}; Mattila et al. \cite{mattila}). In particular,
205: searching the ESO-VLT Archive\footnote{http://archive.eso.org/}, we
206: have found one object, namely the peculiar Type Ia SN~2000cx (Li et
207: al. \cite{li}), that has been observed with VLT-UVES by two different
208: groups, covering two pre-maximum epochs (discussed by Lundqvist et
209: al. \cite{lundqvist03}) and two additional epochs at about two months
210: past maximum (unpublished, see next section). SN~2000cx was discovered
211: about ten days before maximum in the S0 galaxy NGC~524 (Yu, Modjaz \&
212: Li \cite{yu}) and was classified as Type Ia based on its similarity to
213: SN~1991T (Chornock et al. \cite{chornock}). SN~2000cx may be as
214: luminous as SN~1991T (see for example Mazzali et al. \cite{zorro}),
215: but it does not obey to the luminosity vs. light curve width relation
216: (Li et al. \cite{li}) and it is less peculiar from a spectroscopic
217: point of view, since it does not show a Fe lines dominated spectrum at
218: the earliest phases (Filippenko et al. \cite{flipper92}; Ruiz-Lapuente
219: et al. \cite{lapolenta}; Mazzali, Danziger \& Turatto
220: \cite{mazzali95}).
221:
222: Multi-epoch photometry, low-resolution optical and near IR
223: spectroscopy for SN~2000cx have been presented by Li et
224: al. (\cite{li}), Rudy et al. (\cite{rudy}), Sollerman et
225: al. (\cite{sollerman04}) and Branch et al. (\cite{branch04}). Here we
226: focus on the analysis of the high-resolution spectroscopic data
227: obtained for SN~2000cx on four epochs, ranging from day $-$7 to day
228: +68 with respect to $B$ maximum light.
229:
230: The paper is organized as follows. In Sec.~\ref{sec:obs} we describe
231: the observations and the data reduction, while in
232: Sec.~\ref{sec:ngc524} we illustrate the properties of the host
233: galaxy. The results are presented in Sec.~\ref{sec:hires} and
234: discussed in Sec.~\ref{sec:disc}. Finally, in Sec.~\ref{sec:concl} we
235: summarize our conclusions.
236:
237: \section{Observations and Data Reduction}
238: \label{sec:obs}
239:
240: The high-resolution data were obtained with the ESO 8.2m Kueyen
241: Telescope equipped with the Ultraviolet and Visual Echelle
242: Spectrograph (UVES; Dekker et al. \cite{dekker}) on several epochs,
243: using different instrumental setups (DIC1 390+564, DIC2 437+860, RED
244: 580). For our purposes we have selected DIC1 390+564 and RED 580,
245: retaining only the data sets obtained on 2000-07-20, 2000-07-25,
246: 2000-09-17 and 2000-10-03. The other epochs (2000-09-12, 2000-09-13
247: and 2000-09-14) were affected by bad weather and did not add
248: significantly to the phase coverage. The main information on the data
249: is recapped in Table~\ref{tab:obs}.
250:
251: \begin{table}
252: \centerline{
253: \tabcolsep 1.7mm
254: \begin{tabular}{cccccc}
255: \hline
256: Date & Phase & Setup & Range & Exp. Time & Slit Width\\
257: (UT) & (days)& & (\AA) & (seconds) & (arcsec)\\
258: \hline
259: & & & & & \\
260: 2000-07-20 & $-$7 & DIC1 & 3280-4510 & 2$\times$2400 & 0.8\\
261: & & & 4620-5600 & \\
262: & & & 5675-6650 & \\
263: 2000-07-25 & $-$2 & DIC1 & - & 3$\times$3600 & 0.8 \\
264: 2000-09-17 & +52 & RED & 4780-5760 & 2$\times$5400 & 1.0 \\
265: & & & 5835-6810 & & \\
266: 2000-10-03 & +68 & RED & - & 2$\times$5400 & 1.0 \\
267: \hline
268: \end{tabular}
269: }
270: \caption{\label{tab:obs} VLT-UVES observations of SN~2000cx used in this
271: paper. Phase refers to $B$-band maximum light (July 26.7, 2000, Li et
272: al. \cite{li}).}
273: \end{table}
274:
275:
276: The data were reduced using the UVES Data Reduction Pipeline
277: (Ballester et al. \cite{ballester}). Wavelength calibration was
278: achieved using Th-Ar lamps, with a final RMS accuracy of 0.15
279: km s$^{-1}$. The wavelength scale was corrected to the rest-frame
280: adopting a host galaxy heliocentric recession velocity of 2353 km
281: s$^{-1}$ (Emsellem et al. \cite{emsellem}). To compensate for the
282: Earth's motion, a heliocentric velocity correction was applied to
283: the data. The atmospheric lines were identified using a
284: spectroscopically featureless bright star (HR~3239) observed with the
285: same instrumental setup as for the science data. The resolving power
286: $\Delta\lambda/\lambda$ at 5900\AA\/ is 54,000 (DIC1) and
287: 42,600 (RED), corresponding to a full width half maximum (FWHM)
288: resolution of 5.5 and 7.0 km s$^{-1}$ for the two setups,
289: respectively. Finally, in order to increase the signal-to-noise ratio,
290: spectra obtained on the same night were combined. The resulting
291: signal-to-noise ratios per pixel (0.017\AA) in the Na~I D region are
292: 90, 160, 70 and 50 for day $-7$, $-$2, +52 and +68, respectively.
293:
294: The gas column densities relative to the detected absorptions were
295: estimated using VPGUESS\footnote{ VPGUESS has been developed by
296: J. Liske and can be freely downloaded at
297: http://www.eso.org/$\sim$jliske/vpguess/index.html} and
298: VPFIT\footnote{VPFIT has been developed by R.F. Carswell and can be
299: freely downloaded at http://www.ast.cam.ac.uk/$\sim$rfc/vpfit.html}.
300:
301:
302: \begin{figure}
303: \centerline{
304: \includegraphics[width=9cm]{mw.jpg}
305: }
306: \caption{\label{fig:mw} Line profiles for the Galactic Ca~II H\&K
307: and NaI~D lines on day $-$2. The resolution is 5.5 km s$^{-1}$ (FWHM).
308: the vertical tickmarks on the Ca~II K diagram indicate the positions
309: of the main components.}
310: \end{figure}
311:
312:
313: \section{Host galaxy properties and SN location}
314: \label{sec:ngc524}
315:
316:
317: NGC~524 is classified as S0$^+$(rs) (de Vaucouleurs et
318: al. \cite{devauc}), and it shows a gaseous disk with some hint of
319: spiral pattern in the nuclear region (Sarzi et al. \cite{sarzi}).
320: Considering its low ellipticity ($\epsilon<$0.05, Simien \& Prugniel
321: \cite{simien}), it shows very high velocities, with an amplitude that
322: reaches $\sim$140 km s$^{-1}$ at $r \simeq 20$\,\arcsec (Emsellem et
323: al. \cite{emsellem}). The metallicity index Mg\,$b$\/ presents regular
324: gradients as in the case of the H$\beta$ map, which shows a mild
325: positive gradient (Kuntschner et al. \cite{kunt}). The metallicity in
326: the central part of the galaxy is estimated to be [Z/H]$\sim$0.15,
327: while it decreases by about $-$0.35 per dex in $\log (r/r_{eff})$
328: (H. Kuntschner, private communication).
329:
330: SN~2000cx exploded at a projected distance of 111.7 arcsec from the
331: nucleus of NGC~524 (Yu et al. \cite{yu}), corresponding to $\sim$2.3
332: effective radii (Kuntschner et al. \cite{kunt}). At a distance of 33
333: Mpc ($H_0$=72 km s$^{-1}$ Mpc$^{-1}$), this corresponds to about 18
334: kpc. The velocity maps of NGC~524 (Emsellem et al. \cite{emsellem})
335: extend to about 35$^{\prime\prime}$ from the nucleus in the SW
336: direction (where they reach a velocity of about $-$140 km s$^{-1}$)
337: and therefore do not include the apparent position of the explosion
338: site. Nevertheless, these maps suggest that the SN is projected onto
339: the approaching side of the host galaxy. Preliminary results from GMOS
340: observations of NGC524 reaching out to about 2.5 times the effective
341: radius are consistent with the extrapolation of the SAURON velocity
342: maps, albeit indicating velocities of only $\sim-$100 km s$^{-1}$ at
343: the position of SN~2000cx (M. Norris, private communication).
344:
345: \section{Absorption features}
346: \label{sec:hires}
347:
348:
349: \subsection{Milky Way}
350: \label{sec:mw}
351:
352: The Milky Way reddening along the line of sight to NGC~524 is
353: $A_B$=0.356 ($E(B-V)$=0.083, Schlegel, Finkbeiner \& Davis
354: \cite{schlegel}) so that narrow absorption features are expected to
355: be detectable. Indeed, the UVES data we present here clearly show
356: strong Ca~II H\&K and Na~I D lines at heliocentric velocities of 0-30
357: km s$^{-1}$ (see Fig.~\ref{fig:mw}). The global equivalent
358: widths ($EW$) of these lines are 0.24$\pm$0.01 \AA\/ (K),
359: 0.14$\pm$0.01 \AA\/ (H), 0.35$\pm$0.01 \AA\/ (D$_2$) and 0.29$\pm$0.01
360: \AA\/ (D$_1$), respectively. These features show no significant
361: variation during the epochs covered by our data (see
362: Table~\ref{tab:ew}). The ratio of the color excess and the total
363: equivalent width of Na~I D lines is $E(B-V)/EW(NaI D)$=0.13, placing
364: SN~2000cx on the lower slope relation between $EW$ and $E(B-V)$
365: presented by Turatto, Benetti
366: \& Cappellaro (\cite{turatto}).
367:
368:
369: \begin{table}
370: \centerline{
371: \begin{tabular}{ccccc}
372: \hline
373: Epoch & Ca~II K & Ca~II H & Na~I D$_2$ & Na~I D$_1$\\
374: (days)& (\AA) & (\AA) & (\AA) & (\AA)\\
375: \hline
376: $-$7 & 0.24$\pm$0.02 & 0.14$\pm$0.02 & 0.36$\pm$0.03 & 0.29$\pm$0.03\\
377: $-$2 & 0.24$\pm$0.01 & 0.14$\pm$0.01 & 0.35$\pm$0.01 & 0.29$\pm$0.01\\
378: +52 & - & - & 0.35$\pm$0.02 & 0.29$\pm$0.02\\
379: +68 & - & - & 0.36$\pm$0.03 & 0.29$\pm$0.03\\
380: \hline
381: \end{tabular}
382: }
383: \caption{\label{tab:ew} Equivalent widths of Ca~II and Na~I Galactic
384: features.}
385: \end{table}
386:
387: Closer inspection of the line profiles shows a number of
388: sub-components, better seen in the less saturated Ca~II H\&K
389: lines. The main features correspond to +13.5, $-$2.4, $-$7.5, $-$17.7,
390: $-$23.1 and $-$28.5 km s$^{-1}$ (see Fig.~\ref{fig:mw}), the most
391: intense being the second and the third component. The total column density
392: deduced from the Voigt profile fitting of the D$_1$ component gives
393: $\log N(NaI)$=13.1$\pm$0.1. For a typical Milky Way dust mixture this
394: would imply $E(B-V)$=0.24$\pm$0.03 (Hobbs
395: \cite{hobbs}), which is almost a factor 3 larger than the value
396: reported by Schlegel et al.\ (\cite{schlegel}). Even though the column
397: density estimate is certainly hampered by line saturation
398: ($EW(D_2)/EW(D_1)\sim$1.2), we notice that the column density
399: corresponding to $E(B-V)$=0.083 is $\log N(NaI)$=12.3, which can
400: hardly account for the observed line depth.
401:
402:
403: \subsection{Host Galaxy}
404: \label{sec:host}
405:
406:
407: No trace of Na~I absorptions at velocities near the recession velocity
408: of the host galaxy is present in the data (Fig.~\ref{fig:host}, left
409: panels). All the weak absorption features visible in the spectra are
410: in fact identified as telluric lines, with typical $EW$s between 2 and
411: 5 m\AA. An upper limit to the $EW$s of the Na~I D interstellar
412: features is $EW\lesssim$ 1 m\AA. Therefore, a very low amount of
413: interstellar material is expected along the line of sight within the
414: host galaxy. In fact, using the data for the second epoch, when the
415: signal-to-noise is maximum ($\sim$160), we have estimated a 3$\sigma$
416: upper limit for the Na~I column density
417: $N(Na~I)\leq$2$\times$10$^{10}$ cm$^{-2}$ for a velocity parameter
418: $b\leq$10 km s$^{-1}$.
419:
420: A very weak Ca~II absorption is detected at $v$=+176.3 km s$^{-1}$ on
421: both $-$2 and $-$7 days (Fig.~\ref{fig:hosthk}). The column density
422: estimated from the K component on day $-$2 ($EW\sim$4 m\AA) is
423: $N(Ca~II)$=(4.4 $\pm$ 0.5) $\times$10$^{10}$ cm$^{-2}$
424: ($b$=2.1$\pm$1.0 km s$^{-1}$). Since the upper limit for a Na~I D
425: component with the same velocity and velocity parameter on day $-$2 is
426: $N(Na~I)\leq$5$\times$10$^{9}$ cm$^{-2}$, the Na~I to Ca~II ratio is
427: $N(Na~I)/N(Ca~II)\leq$0.1. This is a rather low value, observed in a
428: few cases along the line of sight to some Galactic stars and is
429: usually interpreted as arising in low-density, non-molecular clouds,
430: where calcium depletion onto dust grains is negligible (see Crawford
431: \cite{crawford} and references therein).
432:
433: Since the SN appears to be projected onto the approaching side of
434: NGC~524 (see Sec.~\ref{sec:ngc524}), the high positive velocity
435: measured for the weak Ca~II absorption cannot be explained in terms of
436: galaxy rotation, leading us to conclude that this feature arises in a
437: high-velocity cloud (Wakker \& van Woerden \cite{wakker}), possibly
438: falling into NGC~524, and not related to the circumstellar environment
439: of SN~2000cx.
440:
441:
442: \begin{figure*}
443: \centerline{
444: \includegraphics[width=14cm]{host.jpg}
445: }
446: \caption{\label{fig:host} The Na~I D spectral region for SN~2000cx (left)
447: and SN~2006X (right; P07) at four different epochs. The velocity scale
448: is relative to the Na~I D$_2$ component. The light-colored spectrum in
449: the center of the right and left top panels traces the atmospheric
450: absorption features in the relevant wavelength ranges.}
451: \end{figure*}
452:
453: The absence of narrow emission lines, which would signal
454: interaction between the fast moving SN ejecta and the slow
455: circumstellar wind, was discussed by Lundqvist et
456: al.\ (\cite{lundqvist03}) for the two early epochs. We confirm that
457: there is no trace of H$\alpha$ or He~I 5876\AA\/ in the two later
458: epochs either.
459:
460: \section{Discussion}
461: \label{sec:disc}
462:
463: SN~2000cx exploded in the outskirts of NGC~524. While the
464: non-detection of interstellar lines is compatible with the peripheral
465: position of the SN within its host galaxy, it also marks a difference
466: with respect to the time-variable Na~I D features detected in the
467: normal Type Ia SN~2006X (Fig.~\ref{fig:host}, right panels; P07). In
468: SN~2006X, in fact, while the variable features were barely visible
469: around maximum, they were well developed two weeks later and still
470: clearly present two months later. Therefore, if circumstellar gas was
471: located close to SN~2000cx as in SN~2006X, one might have expected no
472: detectable absorption at the first two epochs (days $-$7 and $-$2),
473: but significant absorption at the two later epochs.
474:
475: The column density deduced from the most intense time-variable feature
476: in SN~2006X (day +14) is $N(NaI)\sim$10$^{12}$ cm$^{-2}$ (P07). If the
477: absorbing material is confined within a thin spherical shell with a
478: radius of $\sim$10$^{16}$ cm, then for solar abundance ($\log
479: N(Na)/N(H)$=$-$5.83; Asplund, Grevesse \& Sauval \cite{asplund}) the
480: implied hydrogen mass is $M(H)\sim$ 10$^{-6}$ M$_\odot$. The upper
481: limit set by our analysis for SN~2000cx on day +52 ($N(NaI)\leq$
482: 5$\times$10$^{10}$ cm $^{-2}$) imposes that in this SN, everything
483: else being the same, $M(H)\leq$ 4$\times$10$^{-8}$ M$_\odot$
484: ($N(H)\leq$ 3$\times$10$^{16}$ cm$^{-2}$ for a solar Na/H abundance
485: ratio).
486:
487: The firm upper limit suggests that the circumstellar environment of
488: SN~2000cx is indeed very different from that of the normal Ia
489: SN~2006X. This can be due to a different {\it i)} progenitor system,
490: {\it ii)} viewing angle, {\it iii)} metallicity or {\it iv)} UV
491: radiation field.
492:
493: \subsection{Progenitor System}
494:
495: SN~2000cx has been classified as a peculiar, if not unique SN Ia (Li
496: et al. \cite{li}). Therefore, one might attribute the lack of Na~I D
497: absorption detection to a different progenitor system, where the
498: circumstellar environment is significantly less dense than in the case
499: of the normal SN~2006X as one would predict, for instance, for the
500: case of a double degenerate scenario. Following the prescriptions
501: given in Benetti et al.\ (\cite{benetti05}), the main
502: spectrophotometric parameters characterizing SN~2000cx are as
503: follows\footnote{The spectroscopic parameters have been derived using
504: unpublished spectra of SN~2006X from the Asiago-ESO-TNG supernova
505: archive.}: $\Delta m_{15}$(B)$_{true}$=0.94 (Li et al. \cite{li} and
506: assuming only Galactic reddening); $\dot{v}$=2 km s$^{-1}$\,d$^{-1}$;
507: $\cal R$(SiII) = 0.11; $v_{10}$(SiII) = 11.85 km s$^{-1}$1000$^{-1}$.
508: These parameters definitely place SN~2000cx in the low-velocity
509: gradient (LVG) subgroup, which includes SN~1991T and all slowly
510: evolving (both photometrically and kinematically) normal SNe~Ia
511: (Benetti et al. \cite{benetti05}). \\ A detailed analysis has shown
512: that this SN is most likely at the edge of normal events, bearing some
513: resemblance to SN~1991T (Mazzali et al. \cite{zorro}). A double
514: degenerate, super-Chandrasekhar scenario has been suggested for both
515: SN~1991T (Fisher et al. \cite{fisher}) and SN~2000cx (Li et al.
516: \cite{li}). In this respect, we emphasize that while SN~1991T, like
517: other over-luminous SNe Ia, was probably associated with a young
518: stellar population (Hamuy et al. \cite{hamuy96}; Howell
519: \cite{howell}) this is not the case for SN~2000cx, which exploded in
520: the outer halo of an S0 Galaxy (see Sec.~\ref{sec:ngc524} and the
521: discussion below). Even though this is certainly not a stringent
522: argument, it suggests a different evolutionary path in the two cases,
523: which do display several dissimilarities (Mazzali et al.
524: \cite{mazzali95}; Li et al. \cite{li}).
525:
526: However, the non-detection of CSM in SN~2000cx does not rule out a
527: more standard single-degenerate scenario. Theoretical models suggest
528: that there are at least two single-degenerate channels that can
529: produce SNe Ia, where the companion is either a main-sequence
530: star/subgiant (the so-called `supersoft channel') or a giant
531: (e.g. Hachisu, Kato \& Nomoto 1999). P07 suggested that the companion
532: in SN~2006X was most likely linked to the giant channel. On the other
533: hand, SN~2000cx may be representative of the supersoft channel (van
534: den Heuvel et al. \cite{vdheuvel}; Rappaport, Di Stefano \& Smith
535: \cite{rappaport}; Han \& Podsiadlowski \cite{han}). In this case, one
536: would expect a lower mass-loss rate from the system and a higher
537: velocity of the wind material, producing a much lower-density
538: circumstellar medium. Moreover, P07 speculated that the density
539: enhancements seen in SN~2006X could be shells of wind material that
540: were swept up by recurrent novae (e.g. Wood-Vasey \& Sokolosky
541: \cite{wood-vasey}). Without such density enhancements the Na~I D
542: recombination time would be very long and the absorption features
543: would be hard to detect. Hence SN~2000cx and SN~2006X could easily
544: both have been produced through different single-degenerate
545: evolutionary channels (also see Parthasarathy et al. \cite{partha} and
546: Tutukov \& Fedorova \cite{tutukov} for recent reviews).
547:
548: In this context, Wang et al. (\cite{wangx07}), based on the
549: photometric and spectroscopic analysis of SN2006X, propose that this
550: object belongs to a subclass of Type Ia events, characterized by
551: distinct properties like very high expansion velocities and abnormal
552: extinction laws.
553:
554:
555:
556: \begin{figure}
557: \centerline{
558: \includegraphics[width=9cm]{hosthk.jpg}
559: }
560: \caption{\label{fig:hosthk}The Ca~II H (right) and K (left) spectral regions
561: for SN~2000cx on days $-$7 (top) and $-$2 (bottom). The vertical
562: dashed line is placed at +176.3 km s$^{-1}$.}
563: \end{figure}
564:
565: \subsection{CSM Geometry}
566:
567: The analysis of P07 suggested that, if the progenitor of SN~2006X was
568: a recurrent nova, the circumstellar structure around similar SNe Ia
569: might only be observed for viewing angles close to the orbital
570: plane. If the material was concentrated in an equatorial torus, it
571: would be easier to decelerate the nova shells and to match the
572: observed velocities. Such an aspherical geometry around recurrent
573: novae is supported by observations of RS Ophiuchi (O'Brien et al.
574: \cite{obrien}; Bode et al. \cite{bode}). In SN~2006X we may have
575: observed the supernova through this equatorial torus, which would help
576: to explain the low observed velocities of the circumstellar shells
577: (P07). However, in general it is more likely that our line of sight
578: does not pass through such a torus. In the model of Bode et al.
579: (2007), the narrow waist covers less than a sixth of the solid angle
580: around the binary. Thus, if the Na~I D lines in SN~2006X
581: were only seen because of a favourable inclination, this would imply
582: that it should only be detectable in a fraction of all SNe Ia (1 in 6
583: SNe in the Bode et al. model for RS Oph).
584:
585: If the material observed in SN~2006X is simply the wind from a
586: subgiant star, it is not clear whether that stellar wind could be
587: aspherical enough for some of the possible lines-of-sight to avoid
588: passing through it.
589:
590:
591: \subsection{Metallicity}
592:
593: An alternative explanation for the lack of detectable Na~I D lines in
594: SN~2000cx is the effect of low metallicity in the accreted material.
595: SN2000cx exploded in the outer regions of NGC~524, at a projected
596: distance of about 18 kpc from the nucleus
597: (Sec.~\ref{sec:ngc524}). Given the abundance gradients observed in
598: early-type galaxies (Carollo, Danziger \& Buson, \cite{carollo}), this
599: most likely implies that the SN was produced by the explosion of a low
600: metallicity star, as proposed by Li et al. (\cite{li}) as a possible
601: explanation for the observed peculiarities. In fact, a metallicity
602: [Na/H]=$-$1.2 would be already sufficient to make SN~2006X-like
603: features undetectable at the last two epochs (+52, +68). Such
604: sub-solar values are not rare in the outer halos of S0 galaxies (see
605: for instance Harris \& Harris \cite{harris}) even though it is not
606: clear that this is the case for NGC~524 (see Sec.~\ref{sec:ngc524}).
607:
608: Even though clear relations seem to exist between the SN properties
609: and the parent population (Filippenko \cite{flipper}; Branch \& van
610: den Bergh \cite{branch94}; Hamuy et al. \cite{hamuy95}; Branch,
611: Romanishin \& Baron \cite{branch96}; Howell \cite{howell}), the origin
612: of this link seems to be age rather than metallicity (Hamuy et
613: al. \cite{hamuy00}; Ivanov, Hamuy \& Pinto \cite{ivanov}). Therefore,
614: a possible low metallicity in the outskirts of NGC~524 might be
615: responsible for the absence of Na~I D circumstellar absorption but, in
616: the light of the currently available data, it is most unlikely that
617: this is the cause of the peculiarities shown by SN~2000cx (Li et
618: al. \cite{li}). In this context, although the models of Kobayashi et
619: al.\ (\cite{koba98, koba06}) predict that no SNe Ia should occur for
620: [Fe/H] $\lesssim$$-$1 within single-degenerate systems, some
621: over-luminous SNe are observed in low metallicity environments. For
622: example, the 1991T-like SN~2007bk (Prieto et al. \cite{prieto2}) went
623: off in the outskirts of its dwarf host galaxy. Prieto et
624: al.\ (\cite{prieto2}) have suggested that this might be an indication
625: that they come from a distinct class of progenitors, such as
626: double-degenerate systems.
627:
628: \subsection{UV deficiency}
629:
630: If what we have seen in SN~2006X is a common property of normal type
631: Ia SNe, then the absence of time-variable absorption features in
632: SN~2000cx may suggest that the UV spectrum was different. In fact, in
633: order for the ionized Na~I to recombine on timescales of the order of
634: 10 days, the medium has to have a sufficiently high electron density
635: ($n_e\geq$ 10$^5$ cm$^{-3}$), which must be produced by the partial
636: ionization of H (P07). Since SN~2000cx was a peculiar object, one
637: possibility is that its UV flux was sufficient to ionize Na~I
638: (ionization potential 5.4 eV), but not to ionize H (13.6 eV). Under
639: these circumstances, the sodium recombination time would become very
640: large and no absorption features would appear. However, the
641: ionization potential of Ca~II is rather high (11.9 eV), and under
642: these conditions one would reasonably expect that most calcium is in
643: the form of Ca~II. Therefore some H\&K absorption components should be
644: visible and their intensity should remain constant in time, as
645: observed in SN~2006X (P07). Since this is not the case, it is hard to
646: believe that a UV deficient SN radiation field is responsible for
647: the lack of Na~I lines in SN~2000cx.
648:
649: \section{Conclusions}
650: \label{sec:concl}
651:
652: Clearly, with only two objects studied it is impossible to decide why
653: SN~2000cx did not show the time-variable features displayed by
654: SN~2006X. As we have seen, the reason could be a different progenitor
655: channel, either a different single-degenerate channel or more
656: radically a double-degenerate system, an orientation effect of the CSM
657: or low metallicity. Many more objects need to be observed in order to
658: settle these open issues, requiring high-resolution spectroscopy
659: covering the first weeks after the explosion, when the existence of
660: circumstellar shells can be revealed before they are swept away by the
661: fast moving SN ejecta. For this purpose we have started a dedicated
662: campaign. The first results, obtained for SN~2007af, will be discussed
663: in a forthcoming paper (Simon et al. \cite{simon}).
664:
665: \begin{acknowledgements}
666: This paper is based on observations made with ESO Telescopes at
667: Paranal Observatory under program IDs 65.H-0426 (P.I. P. Lundqvist)
668: and 65.H-0468 (P.I. M. Della Valle). The authors are indebted to
669: J. Liske, for his kind help during the usage of the VPGUESS package,
670: and to H. Kuntscher and M. Norris for providing them with the
671: metallicity and kinematic data for NGC~524.
672:
673: \end{acknowledgements}
674:
675: \begin{thebibliography}{}
676: \bibitem[2006]{aldering} Aldering, G., et al., 2006, ApJ, 650, 510
677: \bibitem[2005]{asplund} Asplund, M., Grevesse, N. \& Sauval, A.J., 2005,
678: ASPC, 336, 25
679: \bibitem[2000]{ballester} Ballester, P., et al., 2000, The Messenger, 101, 31
680: \bibitem[2005]{benetti05} Benetti, S. et al., 2005, ApJ, 623, 1011
681: \bibitem[2006]{benetti06} Benetti, S., et al., 2006, ApJL, 653, L129
682: \bibitem[2007]{bode} Bode, M.F. et al., 2007, ApJL, in press (arXiv:0706.2745)
683: \bibitem[1993]{branch94} Branch, D. \& van den Bergh, S., 1993, AJ, 105, 2231
684: \bibitem[1995]{branch} Branch, D., Livio, M., Yungelson, L.R.,
685: Boffi \& F., Baron, E., 1995, PASP, 107, 1019
686: \bibitem[1996]{branch96} Branch, D., Romanishin, W. \& Baron, E., 1996,
687: ApJ, 465, 73
688: \bibitem[2004]{branch04} Branch, D., et al., 2004, ApJ, 606, 413
689: \bibitem[1993]{carollo} Carollo, C.M., Danziger, I.J. \& Buson, L.,
690: 1993, MNRAS, 265, 553
691: \bibitem[2000]{chornock} Chornock, R., et al., 2000, IAU Circ. 7463
692: \bibitem[1992]{crawford} Crawford, I., 1992, MNRAS, 259, 47
693: \bibitem[2000]{dekker} Dekker, H., et al., 2000, Proc. SPIE, 4008, 534
694: \bibitem[1991]{devauc} de Vaucouleurs, G., et al., 1991,
695: Third Reference Catalogue of Bright Galaxies, Springer Verlag, Berlin
696: \bibitem[2004]{emsellem} Emsellem, E., et al., 2004, MNRAS, 352, 721
697: \bibitem[1989]{flipper} Filippenko, A.V., 1989, PASP, 101, 588
698: \bibitem[1992]{flipper92} Filippenko, A.V., et al., 1992, ApJ, 384, 15
699: \bibitem[1999]{fisher} Fisher, A., Branch, D., Hatano, K. \& Baron, E.,
700: 1999, MNRAS, 304, 67
701: \bibitem[2004]{gerardy}Gerardy, C.L., et al., 2004, ApJ, 607, 391
702: \bibitem[1999]{hachisu} Hachisu, I., Kato, M. \& Nomoto, K., 1999, ApJ, 522, 487
703: \bibitem[2000]{hamuy00} Hamuy, M., et al., 2000, ApJ, 120, 1479
704: \bibitem[1995]{hamuy95} Hamuy, M., et al., 1995, AJ, 109, 1
705: \bibitem[1996]{hamuy96} Hamuy, M., et al., 1996, AJ, 112, 2391
706: \bibitem[2003]{hamuy} Hamuy, M., et al., 2003, Nature, 424, 651
707: \bibitem[2004]{han} Han Z. \& Podsiadlowski Ph., 2004, MNRAS, 350, 1301
708: %\bibitem[1999]{henry} Henry, R.B.C. \& Worthey, G., 1999, PASP, 111, 919
709: \bibitem[2002]{harris} Harris, W.E. \& Harris. G.L.H., 2002, ApJ, 123, 3108
710: \bibitem[1974]{hobbs} Hobbs, L.M., 1974, ApJ, 191, 391
711: \bibitem[2001]{howell} Howell, D.A., 2001, ApJ, 554, L193
712: \bibitem[2006]{immler} Immler, S.I., et al., 2006, ApJ, 648, L119
713: \bibitem[2000]{ivanov} Ivanov, V., Hamuy, M. \& Pinto, P.A., 2000, ApJ, 542,
714: 588
715: \bibitem[1998]{koba98} Kobayashi, C., Tsujimoto, T., Nomoto, K.,
716: Hachisu, I. \& Kato, M., 1998, ApJL, 503, L155
717: \bibitem[2000]{koba06} Kobayashi, C., Umeda, H., Nomoto, K., Tominaga, N., \&
718: Ohkubo, T., 2006, ApJ, 653, 1145
719: \bibitem[2006]{kunt} Kuntschner, H., et al., 2006, MNRAS, 369, 497
720: \bibitem[2001]{li} Li, W., et al., 2001, PASP, 113, 1178
721: \bibitem[2003]{lundqvist03} Lundqvist, P., et al., 2003, in
722: {\it From twilight to highlight: the physics of Supernovae},
723: ed. W. Hillebrandt \& B. Leibundgut (Berlin: Springer), 309
724: \bibitem[2005]{lundqvist05} Lundqvist, P., et al., 2005, in {\it Cosmic
725: Explosions}, ed. J. M. Marcaide, \& K. W. Weiler, CD-ROM version,
726: IAU Coll., 192, 81
727: \bibitem[2005]{mattila} Mattila, S. et al., 2005, A\&A, 443, 649
728: \bibitem[1995]{mazzali95} Mazzali, P.A., Danziger, I.J. \& Turatto, M.,
729: 1995, A\&A, 297, 509
730: \bibitem[2000]{mazzali00} Mazzali, P.A., 2000, A\&A, 363, 705
731: \bibitem[2005]{mazzali05} Mazzali, P.A., et al., 2005, MNRAS, 357, 200
732: \bibitem[2007]{zorro} Mazzali, P.A., R\"opke, F.K., Benetti, S. \&
733: Hillebrandt, W., 2007, Science, 315, 825
734: \bibitem[2006]{obrien} O'Brien, T.J. et al., 2006, Nature, 442, 279
735: \bibitem[2006]{panagia06} Panagia, N., et al., 2006, ApJ, 469, 396
736: %\bibitem[2007]{panagia07} Panagia, N., 2007, in {\it Supernova 1987A:
737: % 20 Years after: Supernovae and Gamma-Ray Bursters},
738: % S. Immler, R. McCray and K.W. Weiler, Eds. (AIP Conf. Proc., 2007),
739: % in press (http://arxiv.org/abs/0704.1666)
740: \bibitem[2007]{partha} Parthasarathy, M., Branch, D., Jeffery, D.J. \&
741: Baron, E., 2007, New Astronomy Reviews, 51, 524
742: \bibitem[2005]{patat05} Patat, F., Baade, D., Wang, L., Taubenberger, S. \&
743: Wheeler, J.C., 2005, IAU Circ. n.~8631
744: \bibitem[2007]{patat} Patat, F., et al., 2007, Science, 317, 924 ({\bf P07})
745: \bibitem[1996]{pauldrach} Pauldrach, W.A. et al., 1996, A\&A, 312, 525
746: \bibitem[1999]{perlmutter} Perlmutter, S. et al., 1999, ApJ, 517, 565
747: %\bibitem[1992]{phillips} Phillips, M.M., et al., 1992, AJ, 103, 1632
748: \bibitem[2007]{prieto} Prieto, J.L., et al., 2007, ApJ, submitted
749: (arXiv:0706.4088)
750: \bibitem[2007]{prieto2} Prieto, J.L., Stanek, K.Z. \& Beacon, J.F., 2007,
751: ApJ, in press, (arXiv:0706.0690)
752: \bibitem[2006]{quimby06} Quimby, R., et al., 2006, ApJ, 636, 400
753: \bibitem[1994]{rappaport} Rappaport S., Di Stefano R. \& Smith J.D.,
754: 1994, ApJ, 426, 692
755: \bibitem[1998]{riess} Riess, A.G. et al., 1998, AJ, 116, 1009
756: \bibitem[2002]{rudy} Rudy, R.J., et al., 2001, ApJ, 565, 413
757: \bibitem[1992]{lapolenta} Ruiz-Lapuente, P., et al., 1992, ApJ, 387, 33
758: \bibitem[2006]{sarzi} Sarzi, M., et al., 2006, MNRAS, 366, 1151
759: \bibitem[1998]{schlegel} Schlegel, D.J., Finkbeiner, D.P. \& Davis, M.,
760: ApJ, 500, 525
761: \bibitem[2000]{simien} Simien, F. \& Prugniel, Ph., 2000, A\&ASS, 145, 263
762: \bibitem[2007]{simon} Simon, J.D., et al., 2007, ApJ, in preparation
763: \bibitem[2005]{soderberg}Soderberg, A.M., 2006, Astron. Tel., 722, 1
764: \bibitem[2004]{sollerman04} Sollerman, J., et al., 2004, A\&A, 428, 555
765: \bibitem[2005]{sollerman05}Sollerman, J., et al., 2005, A\&A, 429, 559
766: \bibitem[2003]{turatto} Turatto, M., Benetti, S. \& Cappellaro, E., 2003, in
767: {\it From twilight to highlight: the physics of Supernovae},
768: ed. W. Hillebrandt \& B. Leibundgut (Berlin: Springer), 200
769: \bibitem[2007]{tutukov} Tutukov, A.V. \& Fedorova, A.V., 2007,
770: Astronomy Reports, 51, 291
771: \bibitem[1992]{vdheuvel} van den Heuvel E.P.J., Bhattacharya D.,
772: Nomoto K., Rappaport S., 1992, A\&A, 262, 97
773: \bibitem[2003]{wang03} Wang, L., et al., 2003, 591, 1110
774: \bibitem[2007]{wangx07} Wang, X., et al., 2007, ApJ, submitted,
775: (arXiv:0708.0140)
776: \bibitem[1997]{wakker} Wakker, B.P. \& van Worden, H., 1997, ARAA, 35, 217
777: \bibitem[1973]{whelan} Whelan, J. \& Iben, I., 1973, ApJ, 186, 1007
778: %\bibitem[2000]{willson} L.A. Willson, L.A., 2000, ARAA, 38, 573
779: \bibitem[2006]{wood-vasey} Wood-Vasey,W.M. \& Sokoloski, J.L., 2006,
780: ApJ, 645, L53
781: \bibitem[2000]{yu} Yu, C., Modjaz, M. \& Li, W.D., 2000, IAU Circ. 7458
782: \end{thebibliography}
783:
784: \end{document}
785:
786:
787: