1: \documentclass[12pt,manuscript]{aastex}
2: %\documentclass[12pt,preprint]{aastex}
3: %\documentclass{aastex}
4: %\usepackage{emulateapj5}
5:
6: %\documentclass{emulateapj}
7: %\documentclass[onecolumn]{emulateapj}
8: %\documentclass[apjl]{emulateapj}
9: %\usepackage{apjfonts}
10: %(see header for other options)
11:
12: % manuscript produces a one-column, double-spaced document:
13: % \documentclass[manuscript]{aastex}
14:
15: \shorttitle{SN 1994I}
16: \shortauthors{Clocchiatti et al.}
17:
18: %\slugcomment{ Draft version September 2, 2007.}
19:
20: \begin{document}
21: %\received{September 2, 2007}
22:
23: \title{Late-Time {\em HST} Photometry of SN 1994I: Hints of Positron Annihilation Energy Deposition \altaffilmark{1}}
24:
25: \author{
26: Alejandro Clocchiatti,\altaffilmark{2}
27: J. Craig Wheeler,\altaffilmark{3}
28: Robert P. Kirshner,\altaffilmark{4}
29: David Branch,\altaffilmark{5}
30: Peter Challis,\altaffilmark{4}
31: Roger A. Chevalier,\altaffilmark{6}
32: Alexei V. Filippenko,\altaffilmark{7}
33: Claes Fransson,\altaffilmark{8}
34: Peter Garnavich,\altaffilmark{9}
35: Bruno Leibundgut,\altaffilmark{10}
36: Nino Panagia,\altaffilmark{11}
37: Mark M. Phillips,\altaffilmark{12}
38: Nicholas B. Suntzeff,\altaffilmark{13}
39: Peter A. H\"oflich,\altaffilmark{14}
40: and
41: Jos\'e Gallardo\altaffilmark{15}
42: }
43:
44: \altaffiltext{1}{Based in part on observations with the NASA/ESA \emph{Hubble
45: Space Telescope,} obtained at the Space Telescope Science Institute,
46: which is operated by the Association of Universities for Research in
47: Astronomy (AURA), Inc., under NASA contract NAS 5-26555. This
48: research is primarily associated with proposal GO--5777.}
49:
50: \altaffiltext{2}{Pontificia Universidad Cat\'{o}lica de Chile,
51: Departamento de Astronom\'{\i}a y Astrof\'{\i}sica,
52: Casilla 306, Santiago 22, Chile; {aclocchi@astro.puc.cl}.}
53: %
54: \altaffiltext{3}{Department of Astronomy, The University of Texas at Austin,
55: Austin, TX, 78712; {wheel@astro.as.utexas.edu}.}
56: %
57: \altaffiltext{4}{Harvard-Smithsonian Center for Astrophysics, 60 Garden
58: Street, Cambridge,
59: MA 02138; {kirshner@cfa.harvard.edu}, {pchallis@cfa.harvard.edu}.}
60: %
61: \altaffiltext{5}{Homer L. Dodge Department of Physics and Astronomy,
62: University of Oklahoma, 440 West Brooks, Room 100, Norman, OK
63: 73019-2061;{branch@phyast.nhn.ou.edu}.}
64: %
65: \altaffiltext{6}{Department of Astronomy, University of Virginia,
66: P.O. Box 400325, Charlottesville, VA 22904-4325;{rac5x@virginia.edu}.}
67: %
68: \altaffiltext{7}{Department of Astronomy, University of California, Berkeley,
69: CA 94720-3411; {alex@astro.berkeley.edu}.}
70: %
71: \altaffiltext{8}{Department of Astronomy, Stockholm University, AlbaNova, SE-10691 Stockholm, Sweden; {fransson@astro.su.se}.}
72: %
73: \altaffiltext{9}{University of Notre Dame, Department of Physics, 225
74: Niewland Science Hall, Notre Dame, IN 46556-5670;
75: {pgarnavi@miranda.phys.nd.edu}}
76: %
77: \altaffiltext{10}{ESO, Karl-Schwarzschild-Strasse 2, Garching,
78: D-85748, Germany; {bleibund@eso.org}.}
79: %
80: \altaffiltext{11}{Space Telescope Science Institute, 3700 San Martin
81: Drive, Baltimore, MD 21218 (Also, Instituto Nazionale di Astrofisica
82: [INAF], Rome, Italy, and Supernova Ltd., Virgin Gorda, British Virgin
83: Islands).; {panagia@stsci.edu}.}
84: %
85: \altaffiltext{12}{Las Campanas Observatory, Casilla 601, La Serena,
86: Chile;{mmp@lco.cl}.}
87: %
88: \altaffiltext{13}{Texas A\&M University, Physics Department, College
89: Station, TX 77843; {suntzeff@physics.tamu.edu}.}
90: %
91: \altaffiltext{14}{
92: Department of Physics, Florida State University, 315 Keen Building
93: Tallahassee, FL 32306-4350; {pah@physics.fsu.edu}.}
94: %
95: \altaffiltext{15}{
96: Centre de Recherche Astronomique, \'Ecole Normale Superieure de Lyon,
97: France;{jose.gallardo@ens-lyon.fr}.}
98:
99:
100: \begin{abstract}
101:
102: We present multicolor {\em Hubble Space Telescope (HST)} WFPC2
103: broadband observations of the Type Ic SN 1994I
104: obtained $\sim$280~d after maximum light.
105: %
106: We measure the brightness of the SN and,
107: relying on the detailed spectroscopic database of SN~1994I, we
108: transform the ground-based photometry obtained at
109: early times to the {\em HST} photometric system,
110: deriving light curves for the WFPC2
111: $F439W$, $F555W$, $F675W$, and $F814W$ passbands that
112: extend from 7~d before to 280~d after maximum.
113: %
114: We use the multicolor photometry to build a
115: quasi-bolometric light curve of SN~1994I, and compare it
116: with similarly constructed light curves of other supernovae.
117: %
118: In doing so, we propose and test a scaling in energy and time
119: that allows for a more meaningful
120: comparison of the exponential tails of different events.
121: %
122: Through comparison with models, we find that the late-time light
123: curve of SN~1994I is consistent with that of spherically symmetric
124: ejecta in homologous expansion, for which the ability to trap
125: the $\gamma$-rays produced by the radioactive decay of
126: $^{56}$\ion{Co}{0} diminishes roughly as the inverse of time squared.
127: %
128: We also find that by the time of the {\em HST} photometry, the
129: light curve was significantly energized by the annihilation of
130: positrons.
131:
132: \end{abstract}
133:
134: %\keywords{supernovae: general ---
135: %supernovae: individual (\objectname{SN 1994I})}
136: \keywords{Supernovae}
137:
138: \section{Introduction}
139:
140: Late-time light curves are an important piece of evidence for
141: understanding supernova (SN) explosions.
142: %
143: Conditioned by the always diminishing optical depth in the ejecta
144: a time is always reached when the energy from radioactive decay is radiated
145: on a time scale short compared with the dynamical time scale.
146: %
147: Radiative equilibrium is established and there is a simple
148: connection between energy produced, trapped, and radiated.
149: %
150: The simplified physics permits a fairly direct connection between
151: the light curve and the dynamical evolution of the expanding
152: ejecta. This gives a more robust handle on some of the
153: basic physical parameters of the explosion.
154: %
155: How fast these late times are reached depend on the mass to energy ratio
156: of the ejecta.
157: %
158: For supernovae (SNe) without massive H envelopes they start a few weeks
159: after maximum light.
160:
161: The main physical processes at play are the emission of
162: $\gamma$-rays and positrons from radioactive decays in the
163: $^{56}$\ion{Ni}{0} $\rightarrow$ $^{56}$\ion{Co}{0}
164: $\rightarrow$ $^{56}$\ion{Fe}{0}
165: chain \citep{cym69}, their interaction with the ejecta, and the
166: spectrum of the radiation produced by the thermalization
167: processes and the radiative transfer in the expanding ejecta.
168: %
169: Some knowledge of the spectrum is important
170: because ideal bolometric light curves are seldom observed,
171: and astrophysicists generally must deal with observations that
172: cover only a fraction of the radiated energy.
173: %
174: The decay of $^{56}$\ion{Ni}{0} into $^{56}$\ion{Co}{0} releases
175: energy in the form of $\gamma$-rays, with an e-folding time of
176: 8.8~d, while the decay of $^{56}$\ion{Co}{0} into $^{56}$\ion{Fe}{0}
177: releases $\sim 96$\% of the energy in $\gamma$-rays and the
178: remaining $\sim 4$\% as positrons, with an e-folding time of
179: 111.3~d \citep{dandt98}.
180: %
181: Given a density structure and a chemical composition for
182: the expanding ejecta, the $\gamma$-ray transport problem is determined.
183: %
184: Positron interaction with the ejecta, on the other hand, strongly
185: depends on the presence, and geometry, of magnetic fields
186: \citep{rlands98}.
187: %
188: These are, and will probably remain, essentially free parameters.
189: %
190: The escape of positrons from Type Ia supernovae (SNe~Ia) may be
191: relevant to the diffuse 511~keV positron annihilation radiation
192: from the Galactic bulge region \citep{kno05}.
193:
194:
195: Among SNe that explode through the gravitational collapse of
196: the core, there are two distinct scenarios.
197: %
198: For SNe~II \citep[see][for a review of SN types]{filip97},
199: associated with explosions of massive stars that
200: have retained most of their \ion{H}{0}/\ion{He}{0}
201: envelopes, and which are endowed with a relatively large
202: mass-to-energy ratio, the ejecta are able first to trap the
203: UV and optical-near IR photons for $\sim$100 days, and then to trap the
204: $\gamma$-rays from radioactive decays very efficiently for hundreds
205: of days \citep[e.g.][]{cetal96a,eetal03}.
206: %
207: Since they trap essentially all of the energy from the radioactive
208: decay of $^{56}$\ion{Co}{0} during the time when both
209: $\gamma$-rays and positrons are being produced, their light
210: curves do not provide an effective diagnostic for the balance
211: of $\gamma$-ray versus positron energy deposition.
212: %
213: Something similar happens for many (though not necessarily
214: all) Type~Ib and Type~IIb explosions.
215:
216:
217: SNe of Type~Ic, on the other hand, associated with the explosion of
218: stars that were massive while on the main sequence but that
219: through individual and/or interactive binary evolution have lost
220: all of the \ion{H}{0} and most of, if not all, the \ion{He}{0}
221: of the outer layers, present more favorable conditions to study
222: this problem.
223: %
224: Their ejecta have a mass, and/or mass-to-energy ratio, small enough
225: to allow a major fraction of the $\gamma$-rays from $^{56}$\ion{Co}{0}
226: to escape, potentially providing a cleaner window into the details
227: of positron energy deposition.
228:
229: The Type Ic SN 1994I in M51 presented a remarkable opportunity
230: to build a useful late-time light curve for ejecta having both a
231: low mass and a low mass-to-energy ratio.
232: %
233: It exploded in a nearby galaxy and thus reached a bright apparent
234: magnitude at maximum.
235: %
236: Being in a host galaxy that is routinely observed by amateur
237: astronomers meant, in addition, that the SN was spotted,
238: reported, and spectroscopically confirmed well before maximum
239: light \citep{petal94,setal94,cetal94}.
240: %
241: Consequently, it received extensive coverage both near maximum
242: light and at later times.
243:
244:
245: An optical through near-infrared (IR) light curve extending $\sim$50~d
246: after maximum light was prepared by \citet{schmidt94}, and a more
247: extensive one, reaching $\sim$120~d after maximum, was presented
248: by \citet{retal96}.
249: %
250: The SN was, in addition, the target of many other programs,
251: including radio and X-ray observations \citep{retal94a,ietal02},
252: and high-resolution spectroscopy of narrow absorption lines
253: produced by significant amounts of foreground interstellar
254: matter in M51 \citep{hetal95}.
255: %
256: Extensive spectroscopic data sets were collected and presented
257: by \citet{fetal95} and \cite{cetal96b}.
258: %
259: Finally, the light curve derived by \citet{schmidt94} was the subject
260: of detailed theoretical studies to set constraints on the physical
261: parameters of the progenitor of SN 1994I and its evolutionary path
262: to explosion \citep{netal94,ietal94,wlyw95}.
263:
264:
265: In this paper, we present {\em Hubble Space Telescope (HST)}
266: multi-wavelength broadband
267: observations of SN~1994I, taken $\sim$280~d after maximum light
268: ($\sim$300 days after explosive nucleosynthesis).
269: %
270: Combining these observations with those of \citet{retal96},
271: we build an optical through near-IR light curve that, via
272: detailed modeling, can be used to set constraints on the
273: ejected mass and its dynamical and thermal evolution.
274: %
275: We introduce the new observations and photometry in
276: \S~\ref{se:observations}.
277: %
278: In \S~\ref{se:merging}
279: we describe the problems that need to be solved in order
280: to combine the ground-based and space-borne observations,
281: as well as the steps we took to construct our optical through
282: near-IR light curve, and in \S~\ref{se:lcurves} we present the
283: results.
284: %
285: Our discussion in \S~\ref{se:discussion} gives some physical
286: interpretation based both on simplified models of the ejecta
287: and $\gamma$-ray radiative transfer, and presents a revision
288: and extension of the detailed model CO21 \citep{ietal94}.
289: %
290: We summarize our conclusions in \S~\ref{se:conclusions}.
291:
292:
293: \section{Late-Time Observations and Photometry} \label{se:observations}
294:
295: The late-time data on SN~1994I were obtained with {\em HST} under
296: the Supernova INtensive Studies (SINS) General Observer Program.
297: %
298: On 1995 Jan. 15 (UT dates are used throughout this paper), about 280~d
299: after maximum light, the WFPC2 on board {\em HST} was used to take
300: five images of the SN in optical and near-IR passbands,
301: totaling 3200~s of exposure time.
302: %
303: The pointing of the telescope placed the SN on the Planetary
304: Camera chip.
305: %
306: Table~\ref{ta:phot} gives additional details of the observations.
307: %
308: The images were reduced using the ``On-The-Fly Reprocessing'' pipeline,
309: provided by the Space Telescope Science Institute (STScI).
310:
311:
312: Careful inspection of the images revealed the SN at the expected
313: position, amidst the complex structure of one of the inner arms of M51.
314: %
315: The right ascension and declination provided by the world coordinate
316: system of the images was within $1''$ of the position found by
317: high-resolution radio interferometry with the Very Large
318: Array \citep{retal94b}.
319: %
320: The identification was confirmed by analysis of two earlier
321: sets of ultraviolet (UV) images also obtained with {\em HST}.
322: %
323: They were obtained with WFPC2, when the SN was close to maximum light,
324: using the $F336W$ passband.
325: %
326: One image, with an exposure time of 180~s, was taken on 1994 April 18,
327: and a set of three images, with a combined exposure time of 1200~s,
328: was obtained on 1994 May 12.
329: %
330: The object identified as SN~1994I appears with very good signal
331: in all images, and its flux diminishes by about a magnitude
332: between the two epochs.
333: %
334: A finder chart showing SN~1994I in the {\em HST} UV images is given
335: in Figure~\ref{fi:finder}, and stamps of the region around the
336: SN in the UV, optical, and near-IR images are given in
337: Figure~\ref{fi:stamps}.
338:
339: The background against which the SN appeared is complex, but the
340: exquisite resolution of PC1 placed the SN more than 15 pixels
341: away from the closest source.
342: %
343: The background light in the immediate vicinity of the SN appears
344: to be relatively flat.
345: %
346: Point spread function (PSF) fitting photometry of the SN
347: was done using HSTPhot \citep{dol00},
348: which is calibrated with the zero points of \citet{hol95}
349: by normalizing the PSF magnitude to aperture photometry with
350: a $0.5''$ radius.
351: %
352: The adopted version of HSTPhot had zero points and charge-transfer
353: efficiency (CTE) correction parameters updated in 2002 October.
354: %
355: The HSTPhot photometry of SN~1994I is given in Table~\ref{ta:phot}.
356:
357:
358: \section{Merging the Datasets and Estimating Extinction} \label{se:merging}
359:
360: If the late-time photometry of SN~1994I presented in the previous
361: section is combined with the ground-based photometry of
362: \citet{retal96}, the result will be a multicolor light curve
363: extending from $\sim 9$~d before to $\sim 280$~d after maximum light.
364: %
365: Very few SNe~Ic have a multicolor light curve extending so long.
366: %
367: Combining the datasets requires transforming the broadband
368: observations between the different photometric systems used,
369: and applying a correction for interstellar extinction.
370:
371: The ground based observations of \citet{retal96} were taken with three
372: different combinations of instrument and telescopes but all of them
373: referred to the standard system of \citet{bessell90}.
374: %
375: We took the data from their Tables 4, 5, and 6, without making
376: distinctions between datasets obtained with different telescopes,
377: and selecting the dates for which all the $B$, $V$, $R_C,$ and $I_C$ bands
378: had been observed.
379: %
380: When more than one datapoint existed for the same day we combined them
381: by a weighted average.
382: %
383: This approach means that systematic differences in the calibration of the SN
384: photometry due to the slightly different realizations of the standard system
385: are interpreted as an uncertainty in the observation.
386: %
387: The selected photometry is given in Table~\ref{ta:groundphot}.
388:
389:
390: \subsection{The Spectral Shape Corrections}
391:
392: \label{sse:s_corrections}
393:
394: Merging the space-borne and ground-based datasets calls for
395: careful consideration of the passbands used.
396: %
397: The differences in the SN flux measured by two different
398: realizations of the same passband, evaluated in magnitudes,
399: have been called ``spectral shape corrections,'' or $S$--corrections
400: \citep{setal02,ketal04,petal04}.
401: %
402: We generalize here the use of this name to the difference between the
403: flux measured by a ground-based passband and its space-borne
404: counterpart.
405:
406: Computing the $S$--corrections requires precise knowledge of
407: the passbands' throughput and, ideally, spectra of SN~1994I at
408: the specific dates of the photometry.
409: %
410: The spectroscopic evolution of SN~1994I was closely
411: followed \citep{fetal95,cetal96b}, although the last spectrum
412: available was taken only $\sim$147~d after maximum light,
413: more than 130~d before the {\em HST} photometry.
414: %
415: Regarding the passbands, those of the {\em HST} WFPC2 are well
416: characterized\footnote{See the URL ftp://ftp.stsci.edu/cdbs
417: maintained by the STScI.}, but three slightly different
418: realizations of the ground-based $BV(RI)_C$ system
419: were used by \citet{retal96}.
420: %
421: Hence, as a model for the ground-based system, we computed
422: the $S$--corrections as if the standard system were that of
423: \citet{bessell90}, the one to which \citet{retal96} attempted
424: to transform.
425: %
426: We estimated an uncertainty for these $S$--corrections
427: by using different realizations of the Bessell passbands
428: at assorted optical/near-IR imagers at the Cerro Tololo
429: Interamerican Observatory and the Kitt Peak National Observatory.
430: %
431: Note that we transformed the curves of \citet{bessell90}
432: from energy-based to photon-based units, and, for the
433: other ground bassed passbands, we factored in a typical
434: CCD quantum efficiency, the spectral response of two
435: aluminum-coated surfaces, and the typical response of the sky
436: telluric absorption bands if needed.
437:
438:
439: With this database of passbands and spectra, we explored whether
440: it was better to transform the photometry from the space-borne
441: to the ground-based system, or vice-versa.
442: %
443: We found that because spectra are available at the times of the
444: ground-based observations, it was preferable to convert the
445: ground-based photometry to the space-based system.
446: %
447: We computed $S$--corrections for the ground-based photometry
448: using the spectra of \citet{fetal95} and \citet{cetal96b},
449: and passbands as described above. We fit smooth polynomials
450: to their time variation and used them to interpolate the
451: $S$--correction and its uncertainty for the \citet{retal96}
452: observations.
453: %
454: The results are included in Table~\ref{ta:groundphot}.
455:
456:
457: \subsection{Interstellar Extinction}
458:
459: \label{sse:extinction}
460:
461: All of the photometric and spectroscopic observations of
462: SN~1994I indicate that it was heavily reddened by interstellar
463: matter in M51.
464: %
465: Different approaches to estimating $A_V$ have included comparison
466: with detailed theoretical models of light curves \citep{ietal94}
467: and spectra \citep{betal96,betal99,metal99}, and high-resolution
468: spectroscopy of the narrow absorption lines produced by the gaseous
469: phases of the interstellar matter \citep{hetal95}.
470: %
471: Values of $A_V$ from 0.9 up to 3.1 have been inferred, with a
472: preferred range roughly between 1 and 2.
473: %
474: We follow \citet{retal96} in adopting $A_V = 1.4 \pm 0.5$ mag.
475:
476: This value of $A_V$ must next be translated into extinction for
477: the different broadband filters.
478: %
479: Typically an interstellar reddening law is adopted and scaled by
480: the value of $A_V$, and the monochromatic fluxes obtained from
481: the photometry are then corrected, applying the extinction value
482: that the adopted law prescribes for the effective wavelength
483: of the passbands.
484: %
485: This approach, however, may lead to systematic errors when
486: applied to objects with strong and variable emission lines.
487: %
488: The effect is typically small, but it is systematic with time,
489: tending to change the overall shape, and/or slope, of a light curve.
490: %
491: Since the extinction affecting SN~1994I is large, the issue
492: merits consideration.
493:
494: Again, the series of spectra of \citet{fetal95} and
495: \citet{cetal96b} allow us to address this issue.
496: %
497: Using these spectra,
498: a typical interstellar extinction law \citep{ccm89}, with $R_V = 3.1$,
499: the transmittance of the passbands described earlier, and typical
500: instrumental sensitivities, it is possible to
501: compute the extinction affecting the spectra and its
502: variation with time.
503: %
504: The estimated extinction values and their
505: uncertainties are included in Table~\ref{ta:groundphot}.
506:
507:
508: The lack of a spectrum of SN~1994I some 280~d after maximum
509: affects the estimates of interstellar extinction at the time, but
510: the effect is small.
511: %
512: Most of the variations that strongly affect the $S$--corrections
513: at late times (changes of the emission lines in the near-IR) have
514: a minor impact on the extinction.
515:
516:
517: We computed the extinction for the passbands of the {\em HST}
518: WFPC2, fit smooth polynomials to the late-time trends, and
519: extrapolated those from the latest observed spectrum out to
520: the latest photometric point.
521: %
522: To account for the uncertainty of the extrapolation, we increased
523: the typical uncertainty of the fits by a factor of three for the
524: last point.
525: %
526: The extinction computed for the {\em HST} observation
527: is included in Table~\ref{ta:phot}.
528:
529: The uncertainties due to the unknown details of the
530: ground-based standard system realization
531: and the lack of a very late-time spectrum
532: are much smaller than the uncertainty of the overall
533: extinction adopted \citep[$\Delta A_V \approx 0.5$ mag according to][]{retal96}.
534: %
535: The changes in the light curves associated with the
536: variation of the broadband extinction with time fall well
537: within $\Delta A_V$ as well.
538: %
539: However, computing the extinction from the evolving spectra
540: and the expected passbands is the correct procedure for
541: nonstellar spectra, and provides light curves that, for a
542: given velue of $A_V$, are free from the systematic effects
543: related to the time variation of the extinction in each passband.
544:
545: Two additional sources of uncertainty should me mentioned. First, the
546: interstellar extinction law may not be the standard one.
547: %
548: What can be said in this case is that, if the interstellar extinction towards
549: SN~1994I follows the functional form of \citet{ccm89} with a different value of
550: $R_V$ the difference will be smaller than the uncertainty in $A_V$, even for the
551: more extreme values of $R_V$ that they consider.
552: %
553: Second, it is possible that at least a fraction of the extinction is produced in the
554: local environment of the supernova. Since this local environment will be
555: eventually affected by the evolution of the ejecta, a change of the foreground
556: extinction with time would be conceivable. With the data available, it is not
557: possible to set quantitative limits to this possibility.
558:
559: \section{The Light Curves} \label{se:lcurves}
560:
561: The resulting light curves of SN~1994I in the space-based system
562: are shown in Figure~\ref{fi:phot}, while the evolution of the photometric
563: colors is given in Figure~\ref{fi:colors}.
564: %
565: The late-time point is consistent with the decrease in brightness and
566: a flattening of the light curve expected from models that fit the post maximum decline (see below), and as suggested by the ground-based data earlier than $\sim$80~d after maximum.
567: %
568: The $(RI)_C$ magnitudes of the last two ground-based observations
569: (at $\sim$115 and $\sim$140~d after maximum), and the $B$ and
570: $V$ magnitude of the last ground-based observation,
571: are not consistent with this trend.
572: %
573: They appear to be slightly overluminous.
574: %
575: The observed colors indicate little or no change of the overall spectral shape since
576: the latest ground-based photometry.
577:
578: With all the photometry transformed into the same system, it is
579: possible to convert the observed single-bandpass light curves into
580: monochromatic fluxes at the effective wavelengths, integrate
581: them over wavelength, and obtain a new version
582: of the ``$BV(RI)_C$'' light curve of SN~1994I\footnote{Even
583: though it was calibrated using the space-borne WFPC2
584: passbands, we prefer to name the light curve using the more
585: compact designation of their equivalent ground-based passbands.}.
586: %
587: We used the zero points for the WFPC2 passbands given by
588: \citet{hol95}, and integrated the monochromatic fluxes using a
589: simple trapezoid rule.
590: %
591: We did not make any assumptions regarding the behavior of the flux
592: beyond the limiting wavelengths of our coverage;
593: flux beyond the limits of the $B$ and $I_C$ passbands was ignored.
594: %
595: The distance of M51 was assumed to be 8.39 Mpc, based on the
596: luminosity function of planetary nebulae \citep{fcj97}.
597:
598: The $BV(RI)_C$ light curve of SN~1994I is shown in
599: Figure~\ref{fi:BVRI_1}, together with similarly constructed
600: light curves of the Type Ic SNe 1990B \citep{cetal01}, 1998bw
601: \citep{petal01,soetal02,cetal07}, and 2002ap \citep{tetal06}.
602: %
603: The latter two have been modeled with ejecta of low
604: mass-to-energy ratio \citep{ietal98,metal02},
605: but fairly large ejected mass, and have also sometimes been
606: called ``hypernovae.''
607: %
608: As a benchmark of absolute brightness and evolution timescales,
609: the $BV(RI)_C$ light curve of the normal Type Ia SN~1992A (N.
610: Suntzeff, 2006, private communication) is also given.
611:
612:
613: \section{Discussion} \label{se:discussion}
614:
615: \subsection{Simple Comparisons} \label{sse:simple}
616:
617: The light curves in Fig.~\ref{fi:BVRI_1} illustrate the
618: wide range that SNe~Ic span in absolute brightness, and
619: show that the diversity cuts across observational borders
620: proposed to classify them, like the supernova or hypernova
621: categories.
622: %
623: Fig.~\ref{fi:BVRI_1} also shows that SN~1994I
624: was fairly luminous reaching more than 50\% of
625: the brightness of normal Type Ia SN.
626: % SN 92A max=0.39829753E+03
627: % SN 94I max=0.21190142E+03
628: % SN 92A max/SN 94I max=0.39829753E+03 / 0.21190142E+03= 0.5320110469495354932
629: %
630: A foreground extinction a few tenths of a magnitude larger,
631: which is consistent with model light
632: curves, spectra, or high-resolution spectra of the narrow
633: \ion{Na}{0}~$D$ absorption lines, would
634: make it as bright as SN 1992A.
635: %
636: The $BV(RI)_C$ light curves also confirm that, similar to
637: the behavior close to maximum light, the time scale for
638: brightness variation of SN~1994I at late times was fast.
639: %
640: The early exponential tail of SN~1994I is faster than that of
641: the other Type~Ic SNe, and comparable to that of the Type Ia
642: SN.
643:
644: Qualitative comparison with the late-time light curves of
645: SNe~1990B, 1998bw, and 2002ap indicates a mismatch between the
646: trend suggested by the last ground-based photometric
647: points and the HST observation.
648: %
649: SN~1994I shows a sharp downward change of slope in the
650: exponential tail after $\sim$140 days, not seen in the other
651: SNe~Ic.
652: %
653: The rate of decay measured from the data between $\sim$80 and
654: $\sim$ 140~d after explosion is 43\% slower than that measured
655: using the points between $\sim$40 and $\sim$80~d, and the change
656: happens very fast, between 80 and 115~d after explosion.
657: %
658: By contrast, the decline slope of SN~1990B is essentially
659: constant at these times, that of SN~1998bw changes by
660: less than $\sim$28\% between earlier than 80~d after explosion
661: and later than 250~d after explosion, and that of SN~2002ap
662: changes by $\sim$7\% in this same time period.
663:
664: If the trend suggested by the last two ground-based photometric
665: points were correct, then something fairly dramatic happens to
666: the $BV(RI)_C$ light curve between $\sim$140 and $\sim$300~d
667: after maximum light.
668: %
669: One possibility is the so-called ``infrared catastrophe'' that
670: shifts energy from the optical/near-IR passbands into the
671: far-IR \citep{fc89,kandf98}, though it is not expected at
672: such relatively early epochs in SNe~Ic.
673: %
674: Another possibility is that the brightness at 300 days after maximum
675: is affected by a different amount of foreground local extinction, as it would
676: happen if the ejecta of SN~1994I formed dust.
677: %
678: This is, however, not consistent with the small, or nill, change in broadband
679: colors.
680:
681: Another possibility is that the latest ground based photometric points are
682: affected by contamination from background light from M51.
683: %
684: \citet{retal96} did image matching and subtraction before
685: performing photometry, but the background of SN~1994I
686: was particularly bright and had a steep gradient over the spatial
687: scales sampled by the PSF of the ground-based images. Thus,
688: residual contamination in the ground-based images could be present.
689:
690: \subsection{Comparison with a Simple Model}
691:
692: When $\gamma$-ray radiative transfer can be approximated by a
693: pure absorption process \citep{SandW84}, the radioactive nickel
694: assumed to be located at the center of the ejecta, accounting
695: for a small fraction of the total mass (i.e. a point source model
696: for the $\gamma$-ray emission), and the SN ejecta modeled
697: by a spherically symmetric
698: distribution of matter in free expansion, the optical depth that
699: extinguishes the $\gamma$-rays decreases with the inverse of time
700: squared \citep[e.g.][]{candw97}.
701: %
702: In this case, the energy deposited by $\gamma$-rays can be
703: approximated by
704: \begin{equation}\label {eq:deposition}
705: E_\gamma \propto \exp{(-{t/\tau_{\rm Ni}})} \left\{1 - \exp{[-({T_0/t})^2]}\right\},
706: \end{equation}
707: \noindent
708: where $\tau_{\rm Ni}$ is the e-folding time for the decay
709: of $^{56}$\ion{Co}{0}, and $T_0$ is the characteristic time
710: for a given ejecta structure, which depends on its mass $M$, energy
711: $E$, density profile, and the absorption cross section per unit
712: mass ($\kappa$) that matter presents to $\gamma$-rays.
713: %
714: For a given density profile, $T_0 \propto (\kappa M^2/E)^{1/2}$.
715: %
716: Soon after maximum light, when radiative equilibrium is established,
717: the above assumptions are expected to hold.
718: %
719: Then, in addition, between 100\% and $\sim$96\% of the
720: radioactive energy is released in $\gamma$-rays, and most of them
721: are trapped in the ejecta.
722: %
723: In these conditions, $E_\gamma$ gives a good representation of the
724: energy available to heat the ejecta and
725: equation \ref{eq:deposition} can be fitted to bolometric light
726: curves to estimate the parameter $T_0$.
727: %
728: At later times, as the fraction of $\gamma$-rays that escape from the
729: ejecta increases, the $\sim$4\% of energy released in the form of
730: positrons needs to be considered.
731: %
732: In what follows, we will consider the two extreme possibilities for
733: thermalization of the positron annihilation energy: either they are
734: completely trapped in the ejecta, or they escape.
735:
736: Assuming that the $BV(RI)_C$ light curve of SN~1994I, between
737: $\sim$40 and $\sim$80~d after explosion, is a good representation
738: of the bolometric light curve, we fit the light curve to determine
739: $T_0 \approx 65$~d, if the energy from the $^{56}$\ion{Co}{0}
740: positrons is completely deposited, or $T_0 \approx 67$~d, if
741: the positrons escape from the ejecta without interacting.
742: %
743: The $BV(RI)_C$ luminosity with these fitted simple models is
744: shown in Figure~\ref{fi:models}.
745:
746:
747: The simple models are very good at fitting the data between
748: $\sim$40 and $\sim$80~d after explosion,
749: suggesting some connection with reality, and they do a
750: reasonably good job at $\sim$300~d.
751: %
752: Since the (bolometric) model implies $\sim$50\% more energy than
753: the $BV(RI)_C$ observation, the data taken at face value would
754: be consistent either with full trapping of positrons
755: from $^{56}$\ion{Co}{0} {\em and} a
756: small bolometric correction at $\sim$300~d,
757: or with a smaller (even null)
758: bolometric correction and a more complex process of
759: positron transfer that traps $\sim$50\% of their total energy.
760: %
761: It should be noted, though, that the simple model without
762: positron trapping falls below the observed light curve.
763: %
764: The simplest interpretation of this observation is that
765: SN~1994I did trap a significant fraction of the positrons.
766:
767:
768: The last two points from ground-based photometry (between $\sim$100
769: and $\sim$120 days after explosion) are missed by both fits.
770: %
771: We note, in addition, that it is not possible to obtain a good
772: fit for the simple models of eq.~\ref{eq:deposition} when all
773: the ground-based observations of the exponential tail (between
774: $\sim$40 and $\sim$120~d after explosion) are included,
775: even if the {\em HST} point at 300~d is excluded.
776: %
777: The problem is the fast change in the slope
778: of the $BV(RI)_C$ light curve between $\sim$80
779: and $\sim$115~d after explosion.
780: %
781: Fitting this break requires including an
782: additional free parameter, like
783: two different values of $T_0$, which
784: could represent a model of two different mass components
785: each of them with different expansion properties.
786: %
787: As a comparison, it is possible to successfully fit
788: eq.~\ref{eq:deposition} to the exponential decay
789: of the light curves of all the other SNe~Ic shown in
790: Fig.~\ref{fi:BVRI_1}\footnote{We note also that eq. \ref{eq:deposition}
791: fails to provide a good fit to the late-time light curve of the
792: Type Ia SN~1992A.}.
793:
794: \subsection{Simple Comparison with Hindsight From the Simple Model}
795: \label{sse:norm}
796:
797: In \S \ref{sse:simple} we compared the derived
798: light curves of several SNe.
799: %
800: Although it is illustrative to see how different SN light
801: curves behave in a plot like that of Fig.~\ref{fi:BVRI_1},
802: it is not completely correct to compare the evolution of
803: different ejecta on a calendar-time basis.
804: %
805: What we would like to do is to plot the light curves using
806: units that are more physically meaningful.
807:
808:
809: From a few tens to a few hundreds of days after explosive
810: nucleosynthesis, the evolution of the bolometric light
811: curves is governed by two main time scales:
812: $\tau_{\rm Co}$, the e-folding time for the radioactive
813: decay of $^{56}$\ion{Co}{0} into $^{56}$\ion{Fe}{0}, and
814: $T_0$, the characteristic time for the decrease of the
815: $\gamma$-ray optical depth (eq. \ref{eq:deposition}).
816: %
817: They are, respectively, the time scale for energy production
818: and the time scale for energy deposition.
819: %
820: This suggests that the natural {\em energy} scale to compare
821: late-time light curves of different SNe is
822: $L_{\rm Bol}(t_n)/\exp{(-{t_n/\tau_{\rm Co}})}$, where $t_n$
823: is measured since nucleosynthesis,
824: and that the natural scale to compare the time evolution of
825: the ejecta is $t_e/T_0$, where $t_e$ should
826: be measured from the start of free expansion.
827: %
828: Since, at the times of interest, the difference between
829: $t_n$ and $t_e$ is negligible, either of them may be used.
830: %
831: We fitted the models of eq. \ref{eq:deposition} to the
832: light curves of all SNe~Ic in Fig.~\ref{fi:BVRI_1}, and
833: then normalized the energy and time axes according to the
834: previous prescription.
835: %
836: The result is shown in Figure~\ref{fi:BVRInorm}.
837:
838:
839: It is reassuring to confirm that such a simple scaling brings
840: some ``order'' to Fig.~\ref{fi:BVRI_1}.
841: %
842: When the energy is normalized by the input $^{56}$\ion{Co}{0}
843: source, and time by the natural scale for $\gamma$-ray
844: deposition (responsible for most of the energy available to
845: the ejecta from soon after maximum until the supply of
846: $^{56}$\ion{Co}{0} is exhausted), all of the late-time light
847: curves follow approximately the same track.
848: %
849: It is possible to bring them into close agreement just
850: by normalizing the total amount of $^{56}$\ion{Ni}{0}
851: ejected by the explosion (i.e. the additive constant
852: in the logarithm of the energy used in Fig. \ref{fi:BVRI_1}).
853:
854:
855: This comparison shows that, in units of the natural time scale
856: of evolution of the $\gamma$-ray optical depth, the late-time point of SN~1994I
857: presented in this paper ($\sim 280$~d after maximum) is
858: even more extreme than those observed for SN~1998bw and SN 2002ap,
859: which had light curves extending beyond $\sim$500~d after maximum.
860: %
861: Also, it shows that, in comparison with Eq. \ref{eq:deposition},
862: the three very late light curves are consistent
863: with a flattening after $\sim 3 T_0$, a time at which the
864: $\gamma$-ray optical depth becomes negligible, and most of
865: the $\gamma$-rays escape into space.
866: %
867: For SN~1994I, the exact time at which the flattening occurs is
868: unknown due to the lack of observations.
869: %
870: However, both at $3 T_0 \approx 200$~d, and at 300~d corresponding
871: to the last observation, a modest fraction of $^{56}$\ion{Co}{0}
872: remains in the ejecta, and the $\sim$4\% of the radiated energy
873: that its decay releases in the positron channel becomes a
874: significant source.
875:
876:
877: Regarding the late-time ground-based photometry, the scaling
878: of Fig.~\ref{fi:BVRInorm} confirms the arguments given before.
879: %
880: The last two points depart from the common locus, and probably
881: overestimate the luminosity of SN~1994I due to background
882: contamination.
883:
884:
885: \subsection{Insights From a Detailed Model}
886:
887: \citet{ietal94} presented a model of SN 1994I which, without
888: fine tuning of the structure, provides a good fit to the early
889: part of the $UBV(RI)_C$ light curve, from $\sim 6$~d before up
890: to $\sim 40$~d after maximum light.
891: %
892: Their preferred configuration is a C$+$O core that would
893: result from the evolution of a massive star in an interacting binary
894: system that underwent two mass-transfer episodes.
895: %
896: Depending on the initial parameters of the binary, at the moment of
897: explosion the pair consists of the presupernova and a neutron
898: star, white dwarf, or low-mass main-sequence companion
899: \citep{netal94}.
900:
901:
902: \citet{ietal94} fitted the light curves of SN~1994I provided
903: by \citet{schmidt94} with the explosion of a star that entered
904: the main sequence with $\sim 15 M_\sun$, became a
905: $\sim 4.0 M_\sun$ He star after the first mass transfer,
906: and a 2.1 $M_\sun$ C$+$O core after the second (model CO21).
907: %
908: A simultaneous fit of the monochromatic and $UBV(RI)_C$ curves
909: required an interstellar reddening of $E(B-V) = 0.45$ mag, and
910: a distance of 6.92 Mpc ($m-M = 29.2$ mag), in good agreement
911: with the values used to determine the $BV(RI)_C$ light curve
912: presented here.
913:
914:
915: Would model CO21 fit the very late-time {\em HST} photometry?
916: %
917: The question is relevant because previous attempts at fitting
918: SN light curves over long time spans have uncovered
919: a problem. When the models are good at fitting the maximum
920: light they tend to be too dim on the tail; conversely,
921: if they are good at fitting the late-time decay, they tend
922: to give a maximum broader and/or brighter than observed.
923: %
924: Solving this problem has typically meant introducing free
925: parameters in the density structure of the model
926: \citep{eandw88,tetal06}.
927:
928: Based on the explosion model CO21 \citep{ietal94},
929: and using the same radiative transfer code,
930: we calculated the monochromatic light curves, but
931: here we extend the computations to 300 days after
932: explosion.
933: %
934: The assumptions and approximations used in the code
935: are described in more detail in \citep{ietal94}
936: and \citep{hetal93}.
937: %
938: A hypothesis relevant to our discussion below,
939: is that all positrons are thermalized locally
940: and cannot escape.
941: %
942: Because of the limitations of the code
943: in the treatment of forbidden lines and their time dependence,
944: we assumed the energy distribution to be unchanged after day
945: 120.
946: %
947: This assumption is consistent with the observations
948: because the color indices scarcely change after about 100~d
949: (Table~\ref{ta:groundphot} and Fig. \ref{fi:colors}).
950: %
951: The assumption prevents us from testing overall migration
952: of flux toward the IR, as may result from the formation of dust.
953: %
954: The evolution of the colors, however, suggest that neither
955: dust formation, nor other change in the foreground extinction
956: by matter in the circumstellar region, do occur in this SN.
957:
958: From the monochromatic fluxes, we calculated the magnitudes
959: in each relevant $HST$ passband.
960: %
961: We checked that the synthetic $F439W$ and $F555W$ values agree
962: with the $B$ and $V$ of \citet{ietal94} within a few hundredths
963: of a magnitude.
964: %
965: The small differences can be attributed to adjustments in
966: the frequency grid and differences in the {\em HST} filter
967: functions and the Bessell system, since $S$-corrections were
968: not applied in this test.
969:
970: From the monochromatic fluxes at the effective wavelength of
971: the passbands, we built a $BV(RI)_C$ light curve using the same
972: procedure as in the observations (see \S \ref{se:lcurves}).
973: %
974: We compare the theoretical light curves of model CO21 with
975: the observations in Fig.~\ref{fi:models}.
976: %
977: Both the bolometric and $BV(RI)_C$ luminosities are plotted.
978: %
979: They are very similar, indicating that the model concentrates
980: most of the energy within the passbands observed.
981: %
982: Overall, the model luminosities agree well with the observations,
983: especially in the post-maximum decay, although some details
984: merit further comment.
985:
986:
987: At maximum, the model results dimmer than the observations, while
988: the same model fits better the $UBV(RI)_C$ in \citet{ietal94}.
989: %
990: One reason for that is the distance.
991: %
992: The observed $BV(RI)_C$ light curve presented here is based on an
993: observed distance of 8.39 Mpc \citep{fcj97}, while, for the same
994: foreground extinction,
995: CO21 is consistent with a distance 1.47 Mpc shorter in \citet{ietal94}.
996: %
997: The shorter distance will help our model $BV(RI)_C$ light curve to fit
998: the maximum better, but will give a brighter than required post maximum
999: decay.
1000: %
1001: This highlights the importance of using extensive datasets when comparing
1002: models to observations.
1003:
1004: The $BV(RI)_C$ light curve of CO21 shown in Fig.~\ref{fi:models}
1005: does fit most of the observed $BV(RI)_C$ light curve of SN~1994I.
1006: %
1007: First, the distance independent width of the observed maximum, and then
1008: the distance dependent brightness of the post-maximum decline and exponential decay.
1009: %
1010: Fitting the peak-to-tail contrast will require some additional fine tunning.
1011:
1012: A possible hint is that the model $UBV(RI)_C$ light curve does a better job
1013: at fitting the peak-to-tail contrast of the $BV(RI)_C$ observations, indicating
1014: that the model spectrum overestimates the fraction of total flux emerging in the
1015: ultraviolet at early times.
1016: %
1017: In any case, the problem of CO21 is opposite to the traditional peak-to-tail
1018: contrast problem: if the maximum of the $BV(RI)_C$ model is arbitrarily shifted to match
1019: the observed one, then the model light curve results brighter than the observed,
1020: not dimmer.
1021: %
1022: Studying the peak-to-tail contrast problem in SN light curves is beyond the scope of this
1023: paper, so we will let the issue rest there.
1024:
1025: The late-time tail of the detailed model, which includes a
1026: realistic approximation to the $\gamma$-ray transfer and
1027: deposition, follows closely the simple deposition function
1028: given by eq. \ref{eq:deposition}, with full trapping of the
1029: positrons from the decay of $^{56}$\ion{Co}{0}.
1030: %
1031: Since CO21 also assumes full trapping of positrons, this is
1032: an indication that the $\gamma$-ray transfer is well represented
1033: by the simple approximations that went into
1034: eq. \ref{eq:deposition}.
1035: %
1036: Hence, the conclusion stated above that about half of the
1037: positrons are able to deposit their energy within the ejecta
1038: is supported by the more elaborate model as well.
1039:
1040:
1041: \section{Conclusions} \label{se:conclusions}
1042:
1043: Through the {\em HST} GO SINS program, we observed SN~1994
1044: with WFPC2 when the SN was about 280~d past maximum light.
1045: %
1046: We did PSF photometry on the resulting images using
1047: HSTPhot \citep{dol00}.
1048:
1049:
1050: We collected published ground-based photometry of SN~1994I,
1051: applied a reddening correction, and transformed the results
1052: to the space-based {HST} WFPC2 photometric system.
1053: %
1054: Both the reddening correction and the photometric transformation
1055: were based on the extensive spectroscopic database available
1056: for the SN \citep{fetal95, cetal96b}.
1057: %
1058: The resulting light curves in the $F439W$, $F555W$, $F675W$,
1059: and $F814W$ passbands span from $\sim$7~d before to $\sim$280~d
1060: after maximum light.
1061:
1062:
1063: Using the zero points and effective wavelengths of \citet{hol95}, we
1064: transformed the multicolor photometry to monochromatic flux, and
1065: then integrated the flux over the available wavelength range to
1066: construct a ``$BV(RI)_C$'' light curve.
1067: %
1068: We compared the $BV(RI)_C$ light curve of SN~1994I with those of
1069: other SNe~Ic, and, based on a simplified model of the ejecta,
1070: proposed a scaling in energy and time that provides a more
1071: insightful comparison of the exponential tails of different SNe.
1072: %
1073: The light curves that reach times larger than $\sim 3 T_0$,
1074: where $T_0$ is the characteristic time for the
1075: decay of the $\gamma$-ray optical depth, show a clear flattening.
1076: %
1077: For rapidly evolving SNe, like SN~1994I, a simple interpretation
1078: of this flattening is that the small percentage of energy that the
1079: decay of $^{56}$\ion{Co}{0} releases in the form of positrons
1080: becomes a relevant thermal energy source for the ejecta.
1081:
1082:
1083: Model CO21, originally presented by \citet{ietal94}, does a
1084: reasonable job at fitting both the $BV(RI)_C$ and monochromatic
1085: light curves at early times.
1086: %
1087: We extended the original light-curve calculations up to day 300.
1088: We note that the absolute $BV(RI)_C$ luminosity agrees well
1089: with the observations of SN 1994I up to about 80~d after maximum
1090: light, but overestimates the flux by $\sim 50$\% at day 300.
1091:
1092:
1093: Models with full positron trapping indicate that some energy
1094: was missed by the observations, but models without positron
1095: trapping indicate an observed excess of energy.
1096: %
1097: The missing (or excess) energy amounts to $\sim$50\% of that
1098: produced by the annihilation of $^{56}$\ion{Co}{0} positrons.
1099: %
1100: Although models with no positron trapping {\em and} a migration
1101: of energy toward the mid-to-far IR (e.g., by dust formation)
1102: cannot be ruled out, a simpler intepretation of these late-time
1103: observations is that the ejecta of SN~1994I trapped $\sim$50\%
1104: of the positrons roughly 300~d after the explosion.
1105:
1106:
1107: Our interpretation of the $BV(RI)_C$ light curve
1108: is consistent with the SN evolving as a freely expanding
1109: envelope whose optical depth to $\gamma$-rays decreases,
1110: very approximately, with the inverse of time squared.
1111: %
1112: Simple models that treat the deposition of $\gamma$-rays
1113: as a pure absorption process, and the more detailed model
1114: CO21, agree on this.
1115: %
1116: SN~1994I does not need a model with two dynamically distinct
1117: components, like the one that \citet{tetal06} propose for
1118: SN~2002ap.
1119:
1120:
1121: We conclude that our very late-time $BV(RI)_C$ photometry
1122: represents most of the energy that the SN produced at
1123: this time, implying that the late-time bolometric correction of
1124: this rapidly evolving SN~Ic was not large.
1125: %
1126: This, in turn, would mean that the ``infrared catastrophe'' in
1127: this SN does not happen until phases later than $\sim$300~d.
1128:
1129:
1130: Although model CO21 was not finely tuned to closely match
1131: the observed light curves of SN~1994I, it helps to show that
1132: the late-time photometry presented here is extreme in terms
1133: of the $\gamma$-ray optical depth, and agrees with the simple
1134: models of eq. \ref{eq:deposition} in that the dominant source
1135: of late-time energy originates in positrons.
1136: %
1137: A model built specifically to represent this SN could be
1138: used to establish a meaningful constraint on the fraction
1139: of positrons actually trapped in the ejecta, and whether
1140: (and when) there is a migration of photons from the optical
1141: and near-IR to the IR region of the spectrum.
1142:
1143: \acknowledgments A.C. thanks the Department of Astronomy of
1144: the University of Texas at Austin for support from a W. M. Cox
1145: Visitor Grant and from the Samuel T. and Fern Yanagisawa
1146: Regents Professorship in Astronomy, and acknowledges the
1147: support of CONICYT, Chile, under grants FONDECYT 1051061 and
1148: FONDAP 15010003.
1149: %
1150: The {\it HST} SINS program was supported by
1151: NASA through grants GO--2563 and GO--5777 from the Space Telescope Science
1152: Institute, which is operated by the Association of Universities
1153: for Research in Astronomy (AURA), Inc., under NASA contract
1154: NAS 5--26555.
1155: %
1156: J.C.W. was supported in part by NASA Grant NNG04GL00G and NSF
1157: Grant AST--0406740. A.V.F. is grateful for the support of
1158: NSF grant AST--0607485.
1159: %
1160: This research made use of the SUSPECT online database
1161: of SN spectra (http://bruford.nhn.ou.edu/$\sim$suspect/).
1162:
1163: \begin{thebibliography}{}
1164: \input bib.tex
1165: \end{thebibliography}
1166:
1167: \clearpage
1168:
1169: \begin{figure}
1170: \epsscale{0.80}
1171: \plotone{f1.eps}
1172: \caption{\label{fi:finder}
1173: {\em HST} image of SN~1994I and its surroundings.
1174: The SN is at the center of the frame, inside the circle.
1175: North is at the top and East is to the left. The width of the
1176: field displayed is about $28''$. The image shown is the
1177: combination of the three images taken on 1994 May 12, with
1178: the $F336W$ passband.
1179: }
1180: \end{figure}
1181:
1182: \clearpage
1183:
1184: \begin{figure}
1185: \epsscale{0.80}
1186: \plotone{f2.eps}
1187: \caption{\label{fi:stamps}
1188: Stamps of the region around SN~1994I in the {\em HST} images.
1189: North is at the top, and East is to the left. The width of each
1190: stamp is $4.6''$.
1191: %
1192: The passbands used are indicated by labels. SN~1994I is at the
1193: center of each stamp.
1194: The top-left stamp corresponds to the $F336W$ exposure taken on
1195: 1994 April 18, while the top-center stamp is an expanded view
1196: of the 1994 May 12 image shown in Figure~\ref{fi:finder}. The
1197: dimming of SN~1994I in the 24 elapsed days is clear.
1198: %
1199: The other images were taken on 1995 Jan. 15
1200: (see Table~\ref{ta:phot}).
1201: }
1202: \end{figure}
1203:
1204: \clearpage
1205:
1206: \begin{figure}
1207: \epsscale{0.80}
1208: \plotone{f3.eps}
1209: \caption{\label{fi:phot}
1210: {\em HST} WFPC2 broadband unreddened photometry of SN~1994I
1211: (the last point, at JD $-$ 2,449,433 $\approx$ 300),
1212: and ground-based $B$, $V$, $R_C$, and $I_C$ photometry
1213: \citep{retal96} unreddened and translated to the WFPC2
1214: passbands.
1215: %
1216: The vertical marks above the lower horizontal axis indicate
1217: the times at which the spectra of \citet{fetal95} and
1218: \citet{cetal96b} were taken.
1219: %
1220: The shift in time corresponds to the explosion date ($JD =
1221: 2,449,433$) estimated by \citet{ietal94}.
1222: }
1223: \end{figure}
1224:
1225: \clearpage
1226:
1227: \begin{figure}
1228: \epsscale{0.80}
1229: \plotone{f4.eps}
1230: \caption{\label{fi:colors}
1231: {\em HST} WFPC2 broadband unreddened colors of SN~1994I
1232: in the HST passbands
1233: computed from the photometry of Figure{\protect{\ref{fi:phot}}}.
1234: %
1235: The shift in time corresponds to the explosion date ($JD =
1236: 2,449,433$) estimated by \citet{ietal94}.
1237: }
1238: \end{figure}
1239:
1240: \clearpage
1241:
1242: \begin{figure}
1243: \epsscale{0.75}
1244: \plotone{f5.eps}
1245: \caption{ \label{fi:BVRI_1}
1246: $BV(RI)_C$ light curves of the Type~Ic SNe~1994I, 1990B,
1247: 1998bw, and 2002ap, as well as of the Type~Ia SN~1992A.
1248: %
1249: The light curve of SN~1990B has been displaced downward by
1250: 0.5 vertical units for clarity.
1251: %
1252: Explosion-date estimates are from a theoretical light-curve
1253: fit for SN~1994I \citep{ietal94}, theoretical spectra and
1254: light-curve fits for SN~2002ap \citep{metal02}, and the date
1255: of GRB 980425 for SN~1998bw.
1256: %
1257: For SN~1990B the explosion date was taken from \citet{cetal01},
1258: who estimated it from comparison with SN~1993J, and for
1259: SN~1992A the explosion date was assumed to be $\sim$20~d before
1260: maximum light in $B$.
1261: %
1262: The slope for full trapping of the $^{56}$\ion{Co}{0} decay
1263: products is also shown.
1264: }
1265: \end{figure}
1266:
1267: \clearpage
1268:
1269: \begin{figure}
1270: \epsscale{0.80}
1271: \plotone{f6.eps}
1272: \caption{ \label{fi:models}
1273: $BV(RI)_C$ light curve of SN~1994I, the simple models given by
1274: eq.~\ref{eq:deposition}, and model CO21.
1275: }
1276: \end{figure}
1277:
1278: \clearpage
1279:
1280: \begin{figure}
1281: \epsscale{0.90}
1282: \plotone{f7.eps}
1283: \caption{ \label{fi:BVRInorm}
1284: $BV(RI)_C$ late-time light curves of SNe~Ic 1994I, 1998bw, 2002ap,
1285: and 1990B, normalized according to the prescriptions of
1286: \S~\ref{sse:norm}. Points mark the observations of SN~1994I.
1287: Solid symbols indicate the region
1288: used to fit the time-scale parameter $T_0$.
1289: %
1290: A constant was added to the light curves of SNe~1994I, 1998bw,
1291: 2002ap, and 1990B, to make them match the solid points
1292: of SN~1994I.
1293: }
1294: \end{figure}
1295:
1296: \clearpage
1297:
1298: \begin{deluxetable}{llrcrr}
1299: \tablewidth{0pt}
1300: \tablecaption{Late-Time {\em HST} Photometry of SN~1994I\label{ta:phot}}
1301: \tablehead{
1302: \colhead{{\em HST} Dataset} &
1303: \colhead{JD\tablenotemark{a}} &
1304: \colhead{Exp.\tablenotemark{b}} &
1305: \colhead{Passband} &
1306: \colhead{{\em HST} Mag.} &
1307: \colhead{Extinction\tablenotemark{c}}
1308: }
1309: \startdata
1310: U2LQ0201T/2T & 733.21 & 1400 & $F439W$ & 21.63 $\pm$ 0.02 & 1.89 $\pm$ 0.033 \hfill \\
1311: U2LQ0203T & 733.22 & 600 & $F555W$ & 21.03 $\pm$ 0.01 & 1.45 $\pm$ 0.030 \hfill \\
1312: U2LQ0204T & 733.25 & 600 & $F675W$ & 20.21 $\pm$ 0.01 & 1.12 $\pm$ 0.010 \hfill \\
1313: U2LQ0205T & 733.26 & 600 & $F814W$ & 19.69 $\pm$ 0.01 & 0.87 $\pm$ 0.010 \hfill \\
1314: \enddata
1315: \tablenotetext{a}{Mean Julian Date of total exposure minus 2,449,000. For
1316: phase, subtract an additional 433.00.}
1317: \tablenotetext{b}{Total exposure time in the passband (in seconds). Two
1318: $F439W$ exposures of 700~s each were combined in one image.}
1319: \tablenotetext{c}{Interstellar extinction in the host galaxy, in the
1320: {\em HST} passband.}
1321: \end{deluxetable}
1322:
1323: \begin{deluxetable}{lllllllllrrrr}
1324: \rotate
1325: \tablewidth{0pt}
1326: \tablecaption{Ground-Based Photometry of SN~1994I\label{ta:groundphot}}
1327: \tablehead{
1328: \colhead{JD\tablenotemark{a}} &
1329: \colhead{$B$\tablenotemark{b}} &
1330: \colhead{$V$\tablenotemark{b}} &
1331: \colhead{$R_C$\tablenotemark{b}} &
1332: \colhead{$I_C$\tablenotemark{b}} &
1333: \colhead{$A_B$\tablenotemark{c}} &
1334: \colhead{$A_V$\tablenotemark{c}} &
1335: \colhead{$A_{R_C}$\tablenotemark{c}} &
1336: \colhead{$A_{I_C}$\tablenotemark{c}} &
1337: \colhead{$SC_B$\tablenotemark{d}} &
1338: \colhead{$SC_V$\tablenotemark{d}} &
1339: \colhead{$SC_{R_C}$\tablenotemark{d}} &
1340: \colhead{$SC_{I_C}$\tablenotemark{d}}
1341: }
1342: \startdata
1343: 445 & 14.41 $\pm$ 0.03 & 14.04 $\pm$ 0.02 & 13.87 $\pm$ 0.02 & 13.77 $\pm$ 0.01 & 1.83 & 1.40 & 1.17 & 0.86 & -0.03 & -0.04 & 0.05 & -0.01 \\
1344: 447 & 13.97 $\pm$ 0.08 & 13.39 $\pm$ 0.02 & 13.21 $\pm$ 0.02 & 13.09 $\pm$ 0.02 & 1.83 & 1.40 & 1.17 & 0.86 & -0.04 & -0.04 & 0.06 & -0.01 \\
1345: 448 & 13.95 $\pm$ 0.02 & 13.20 $\pm$ 0.01 & 13.02 $\pm$ 0.01 & 12.88 $\pm$ 0.01 & 1.83 & 1.40 & 1.17 & 0.86 & -0.05 & -0.04 & 0.06 & -0.01 \\
1346: 450 & 13.68 $\pm$ 0.21 & 12.99 $\pm$ 0.02 & 12.72 $\pm$ 0.02 & 12.60 $\pm$ 0.02 & 1.83 & 1.40 & 1.17 & 0.86 & -0.06 & -0.04 & 0.06 & 0.00 \\
1347: 453 & 13.81 $\pm$ 0.07 & 12.91 $\pm$ 0.02 & 12.62 $\pm$ 0.02 & 12.49 $\pm$ 0.02 & 1.83 & 1.40 & 1.17 & 0.86 & -0.07 & -0.04 & 0.06 & 0.00 \\
1348: 454 & 14.05 $\pm$ 0.02 & 13.02 $\pm$ 0.01 & 12.72 $\pm$ 0.01 & 12.52 $\pm$ 0.01 & 1.83 & 1.40 & 1.17 & 0.86 & -0.07 & -0.04 & 0.06 & 0.00 \\
1349: 455 & 14.15 $\pm$ 0.02 & 13.11 $\pm$ 0.01 & 12.76 $\pm$ 0.01 & 12.53 $\pm$ 0.01 & 1.83 & 1.40 & 1.17 & 0.86 & -0.08 & -0.04 & 0.06 & 0.00 \\
1350: 456 & 14.44 $\pm$ 0.01 & 13.23 $\pm$ 0.01 & 12.86 $\pm$ 0.01 & 12.61 $\pm$ 0.01 & 1.83 & 1.40 & 1.17 & 0.86 & -0.08 & -0.04 & 0.06 & 0.00 \\
1351: 457 & 14.56 $\pm$ 0.02 & 13.34 $\pm$ 0.01 & 12.94 $\pm$ 0.02 & 12.64 $\pm$ 0.01 & 1.83 & 1.40 & 1.17 & 0.86 & -0.08 & -0.04 & 0.06 & 0.00 \\
1352: 458 & 14.75 $\pm$ 0.02 & 13.46 $\pm$ 0.02 & 13.04 $\pm$ 0.02 & 12.72 $\pm$ 0.02 & 1.83 & 1.41 & 1.17 & 0.86 & -0.08 & -0.04 & 0.06 & 0.00 \\
1353: 461 & 15.28 $\pm$ 0.03 & 13.93 $\pm$ 0.02 & 13.41 $\pm$ 0.02 & 13.00 $\pm$ 0.02 & 1.83 & 1.41 & 1.17 & 0.86 & -0.09 & -0.04 & 0.06 & 0.00 \\
1354: 462 & 15.40 $\pm$ 0.04 & 14.08 $\pm$ 0.02 & 13.53 $\pm$ 0.03 & 13.10 $\pm$ 0.02 & 1.83 & 1.41 & 1.17 & 0.85 & -0.09 & -0.03 & 0.06 & 0.00 \\
1355: 470 & 16.00 $\pm$ 0.32 & 14.92 $\pm$ 0.02 & 14.37 $\pm$ 0.03 & 13.79 $\pm$ 0.01 & 1.84 & 1.41 & 1.17 & 0.85 & -0.09 & -0.03 & 0.07 & 0.01 \\
1356: 471 & 15.97 $\pm$ 0.30 & 15.11 $\pm$ 0.02 & 14.50 $\pm$ 0.02 & 13.96 $\pm$ 0.02 & 1.84 & 1.41 & 1.17 & 0.85 & -0.09 & -0.03 & 0.07 & 0.01 \\
1357: 476 & 16.28 $\pm$ 0.10 & 15.41 $\pm$ 0.05 & 14.80 $\pm$ 0.07 & 14.12 $\pm$ 0.01 & 1.84 & 1.41 & 1.17 & 0.85 & -0.08 & -0.02 & 0.07 & 0.01 \\
1358: 482 & 16.61 $\pm$ 0.08 & 15.56 $\pm$ 0.02 & 15.03 $\pm$ 0.03 & 14.37 $\pm$ 0.02 & 1.84 & 1.41 & 1.17 & 0.85 & -0.06 & -0.02 & 0.07 & 0.01 \\
1359: 486 & 16.56 $\pm$ 0.11 & 15.71 $\pm$ 0.03 & 15.20 $\pm$ 0.04 & 14.53 $\pm$ 0.01 & 1.84 & 1.41 & 1.17 & 0.85 & -0.05 & -0.01 & 0.08 & 0.01 \\
1360: 487 & 16.78 $\pm$ 0.04 & 15.76 $\pm$ 0.02 & 15.24 $\pm$ 0.03 & 14.54 $\pm$ 0.02 & 1.84 & 1.41 & 1.17 & 0.85 & -0.05 & -0.01 & 0.08 & 0.01 \\
1361: 491 & 16.58 $\pm$ 0.16 & 15.90 $\pm$ 0.03 & 15.41 $\pm$ 0.07 & 14.62 $\pm$ 0.02 & 1.84 & 1.41 & 1.17 & 0.85 & -0.04 & -0.01 & 0.08 & 0.01 \\
1362: 494 & 16.88 $\pm$ 0.07 & 15.94 $\pm$ 0.02 & 15.45 $\pm$ 0.03 & 14.75 $\pm$ 0.02 & 1.84 & 1.41 & 1.17 & 0.85 & -0.03 & -0.01 & 0.08 & 0.02 \\
1363: 495 & 16.86 $\pm$ 0.09 & 15.89 $\pm$ 0.02 & 15.49 $\pm$ 0.02 & 14.80 $\pm$ 0.03 & 1.84 & 1.41 & 1.17 & 0.85 & -0.03 & -0.01 & 0.08 & 0.02 \\
1364: 502 & 17.21 $\pm$ 0.14 & 16.18 $\pm$ 0.07 & 15.73 $\pm$ 0.09 & 15.00 $\pm$ 0.03 & 1.84 & 1.41 & 1.17 & 0.85 & -0.02 & 0.00 & 0.08 & 0.02 \\
1365: 503 & 17.06 $\pm$ 0.06 & 16.24 $\pm$ 0.02 & 15.75 $\pm$ 0.04 & 15.04 $\pm$ 0.02 & 1.84 & 1.41 & 1.17 & 0.85 & -0.02 & 0.00 & 0.08 & 0.02 \\
1366: 505 & 16.93 $\pm$ 0.17 & 16.21 $\pm$ 0.04 & 15.74 $\pm$ 0.06 & 15.08 $\pm$ 0.02 & 1.84 & 1.41 & 1.17 & 0.85 & -0.01 & 0.00 & 0.09 & 0.02 \\
1367: 512 & 17.21 $\pm$ 0.18 & 16.48 $\pm$ 0.04 & 15.99 $\pm$ 0.06 & 15.30 $\pm$ 0.02 & 1.84 & 1.42 & 1.17 & 0.85 & -0.01 & 0.01 & 0.09 & 0.02 \\
1368: 515 & 17.48 $\pm$ 0.08 & 16.58 $\pm$ 0.04 & 15.97 $\pm$ 0.03 & 15.37 $\pm$ 0.05 & 1.84 & 1.42 & 1.17 & 0.85 & 0.00 & 0.01 & 0.09 & 0.02 \\
1369: 548 & 18.05 $\pm$ 0.13 & 17.23 $\pm$ 0.08 & 16.46 $\pm$ 0.08 & 15.95 $\pm$ 0.05 & 1.85 & 1.42 & 1.17 & 0.86 & 0.00 & 0.03 & 0.11 & 0.03 \\
1370: 572 & 18.10 $\pm$ 0.18 & 17.60 $\pm$ 0.06 & 16.93 $\pm$ 0.09 & 16.34 $\pm$ 0.05 & 1.85 & 1.42 & 1.17 & 0.86 & 0.00 & 0.04 & 0.12 & 0.04 \\
1371: \enddata
1372: \tablenotetext{a}{Mean Julian Date minus 2,449,000. For phase, subtract an
1373: additional 433.00.}
1374: \tablenotetext{b}{Photometry from Tables 4, 5, or 6 of
1375: {\protect{\citet{retal96}}}.}
1376: \tablenotetext{c}{Interstellar extinction in the host galaxy, in the
1377: given ground-based passband.
1378: Mean uncertainties (mag) are
1379: $\delta A_B = 0.060$,
1380: $\delta A_V = 0.010$,
1381: $\delta A_{Rc} = 0.003$, and
1382: $\delta A_{Ic} = 0.003$.
1383: }
1384: \tablenotetext{d}{$S$--correction to be subtracted from the
1385: ground-based magnitude to transform to the space-borne
1386: equivalent passband (see Table \protect{\ref{ta:phot}}).
1387: Mean uncertainties (mag) are
1388: $\delta SC_B = 0.033$,
1389: $\delta SC_B = 0.041$,
1390: $\delta SC_{Rc} = 0.053$, and
1391: $\delta SC_{Ic} = 0.014$.
1392: }
1393: \end{deluxetable}
1394:
1395: \end{document}
1396: