arXiv:1304.4934v2 [astro-ph.HE] 22 Jul 2013 04 leige l 06 reoe l 07.Tedata The 2007). order al. al. (of et loss et Wood-Vasey mass Prieto 2004; substantial 2006; suggest 2004; al. al. al. et et et Wang Aldering Deng 2004; 2004; 2003; al. al. et et Kotak (Hamuy 2005gj and 2002ic 2013). al. has a et scenario) (Soker after proposed core-degenerate merger been prompt a also violent al- (i.e., 2013a), a envelope al. particu- from common et loss proba- this Silverman 2012; mass a least al. though confirm et at (Dilday for PTF11kx event lar channel Ia-CSM single-degenerate SN ble nearby the of agrpa uioiista yia N a nthe in Ia, SNe slightly typical have also than Ia-CSM luminosities SNe range The peak Filippenko review). larger see slow, a 1990; a for relatively (Schlegel 1997 from and CSM strong generated to pre-existing their lines similarities dense, by refer- emission many defined hydrogen and show IIn, narrow spectra 2013b, SNe (here- of The al. medium those et for- Silverman circumstellar within). the Ia-CSM; dense ences between SNe a interaction after and al. significant shock et of ward Bloom signs 2012a,b; hibit 2011; al. et 2013a). al. Foley al. et et Ganeshalingam al. Silverman 2012; et 2012; 2011; Simon al. 2009; et al. al. Brown et et for Nugent Blondin 2007; exists 2009; al. channels now et double-degenerate Patat Evidence compan- (e.g., and the single-degenerate ambiguous. of a nature both remains of gen- the explosion is but thermonuclear ion (WD), itself dwarf the SN white be empirical The underlying C/O to these the accepted systems. erally Despite progenitor about and remain indi- physics 1993). questions distance Phillips cosmological relationships, (e.g., precise yields cators curves light rmtecmainsa n infiataonsof amounts significant and star companion the from rpittpstuigL using 2018 typeset 11, Preprint August version Draft w ftems elsuidSeI-S r SNe are Ia-CSM SNe well-studied most the of Two eetsuisrva usml fSeI htex- that Ia SNe of subsample a reveal studies Recent Ia) (SN Ia Type standardize to ability The 1 2 eateto srnm,Uiest fClfri,Berkel California, of University Astronomy, of Department mi:oo@eklyeu. [email protected] email: − 21 neato SeI-S) epeetupbihdarchival unpublished present We Ia-CSM). (SNe interaction aCM20i n 05jobtained 2005gj and 2002ic Ia-CSM neato.I h aeo N20g,temdI uioiymr th more luminosity optical-wav mid-IR late-time headings: interac the any Subject shock renewed of 2005gj, to aware rebrightening SN not this ge attribute of are continuously t we we case beyond emission epochs, While lies X-ray the that and In post-discovery. shell optical dust by war pre-existing interaction. heated from a radiatively emission with likely (mid-IR) consistent mid-infrared most late-time are for evidence show rwn ubro bevtosrva usto yeI supe Ia Type of subset a reveal observations of number growing A . 3 H AETM ERGTNN FTP AS 05JI H MID-I THE IN 2005GJ SN IA TYPE OF REBRIGHTENING LATE-TIME THE ≤ M R 1. − ≤ A INTRODUCTION T E tl mltajv 5/2/11 v. emulateapj style X 9mg ealdobservations Detailed mag. 19 icmtla atr—sproa:gnrl—sproa:individ supernovae: — general supernovae: — matter circumstellar N20g)—ds,etnto nrrd infrared: — extinction dust, — 2005gj) SN r .Fox D. Ori > 10 − 4 rf eso uut1,2018 11, August version Draft > M ⊙ 30ad50dy otdsoey epciey ohSNe Both respectively. post-discovery, days 500 and 1300 1,2 yr ABSTRACT lxiV Filippenko V. Alexei & − ey 1 ) n oterdneCM N I xii aetm ( late-time Ow- exhibit IIn. IIn SNe SNe CSM, of dense those their to to ing similar These also (near-IR). are near-infrared characteristics the in emitting dust warm rsnsorconclusions. our presents obtained elescope T Space Spitzer mass-loss progenitor the about shells history. dust clues al. these important from et derived Fox reveals geometry 2002; CSM al. The et continuously 2011). Gerardy ongo- is by (e.g., interaction generated that CSM radiation dust ing X-ray and warm visible by from heated emission IR day) i n etn ehns fteds in ori- dust the the explore of observations; We mechanism heating the luminosity. and the of gin thus details and the perature, lists 2 Spitzer Section tively. est h otBscClbae aa( Data ac- Calibrated provide SHA Basic The Post and the 2002ic 1. to Table SNe cess in of summarized observations 2005gj, unpublished revealed e.W hrfr efre pruepooer with photometry aperture performed DAOPHOT therefore We ies. h aatcnces(leige l 06 reoe al. et Prieto http://sha.ipac.caltech.edu/applications/Spitzer/SH 2006; al. et (Aldering nucleus galactic the nrpdsailsae.Tmlt utato a re- can , pre-SN subtraction underlying no the Template but from confusion scales. photometric duce variations spatial host background-flux SN rapid exhibit The and on 5.8 bright epoch. be and single to 4.5, tend a 3.6, at combined images shows the 1 channel Figure of images calibrated. and false-color coadded fully already are AR)udrcoeaieareetwt h NSF. Astr the in ope with Research is agreement for cooperative Universities which under of Observatory, (AURA) Association Astronomy the Optical by ated National the by 4 3 nthis In erhi the in search A RF h mg euto n nlssFclt sdistribu is Facility Analysis and Reduction Image the IRAF: htmtycntan h utms n tem- and mass dust the constrains photometry / 2.1. pte pc elescope T Space Spitzer H a eacse from accessed be can SHA APPHOT > inwt es icmtla shell. circumstellar dense a with tion Letter 30ad50dy otdsoey respec- post-discovery, days 500 and 1300 Warm 1 nveudron circumstellar undergoing rnovae Spitzer efradsokrdu,most radius, forward-shock he ut h utparameters dust The dust. m in lnt bevtosa these at observations elength 2. epeetupbihdarchival unpublished present we Spitzer eae ylt-ieCSM late-time by nerated IRAF Spitzer OBSERVATIONS ndulsatr1year 1 after doubles an epae xs o hs galax- these for exist templates aao N 02cad2005gj and 2002ic SNe on data . 4 IA Photometry /IRAC N20g falls 2005gj SN eiaeAcie(SHA) Archive Heritage a S 2002ic, (SN ual NFRARED aao SNe on data pbcd § .Scin4 Section 3. /. A/ < ,which ), 1 ′′ > onomy from 100 µ ted m r- 3 a=0.1 µm 2

Table 1 Spitzer Observations1

3 3 3 3 SN JD Epoch PID AOR Distance tint 3.6 µm 4.5 µm 5.8 µm 8.0 µm −2,450,000 (days) (Mpc) (s) (1017 erg s−1 cm−2 A˚−1) 2002ic 3386 795 3248 10550272 280 4000 0.150(0.015) 0.136(0.012) 0.092(0.008) 0.052(0.004) 2002ic 3770 1179 20256 14455040 280 3600 0.060(0.009) 0.063(0.008) 0.052(0.006) 0.040(0.004) 2002ic 3961 1370 30292 17965824 280 2400 0.042(0.008) 0.045(0.007) 0.037(0.005) 0.031(0.003) 2002ic 4356 1765 40619 23107840 280 2400 <0.03 <0.03 <0.03 <0.03 2005gj 3778 139 264 16868096 268 2400 0.130(0.014) 0.092(0.010) 0.036(0.005) 0.017(0.003) 2005gj 4004 365 30733 19308800 268 2400 0.183(0.016) 0.110(0.011) 0.049(0.006) 0.024(0.003) 2005gj 4149 510 30733 19309056 268 2400 0.276(0.020) 0.199(0.014) 0.113(0.009) 0.049(0.004) 1 Upper limits for nondetections were derived by the point-source sensitivity in Table 2.10 of the IRAC Instrument Handbook, version 2. 2 −1 −1 µ All distances are derived from the host-galaxy assuming H0 = 72 km s Mpc . a=0.1 m 3 1σ uncertainties are given in parentheses. SN 2002ic 0.10 ) -2 795 d cm -1 1179 d erg s -12 1370 d (10 λ F λ

0.01 2 4 6 8 λ (µm) (a) SN2002ic (a)

SN 2005gj 0.10 )

-2 510 d cm -1 erg s

-12 365 d (10 λ 139 d F λ

0.01 2 4 6 8 λ (µm) (b) SN2005gj (b) Figure 1. Combined false-color 3.6, 4.5, 5.8 µm images of SNe Figure 2. Photometry of SNe 2002ic and 2005gj in Spitzer/IRAC 2002ic and 2005gj at late times (∼1′′×1′′). Epochs refer to days Channels 1 (3.6 µm), 2 (4.5 µm), 3 (5.8 µm), and 4 (8.0 µm). post-discovery. Overplotted are the resulting best fits of Equation 1.

2007). To minimize contributions from the underlying assuming optically thin dust with particle radius a, at galaxy in both cases, a 2-pixel radius was chosen and a distance d from the observer, thermally emitting at a aperture corrections were applied, although removal of single equilibrium temperature (e.g., Hildebrand 1983), nuclear contributions is admittedly difficult. where Bν (Td) is the Planck blackbody function and κν (a) is the dust absorption coefficient. 2.2. Dust Temperature and Mass For simplicity, we assume a simple dust population of a single size composed entirely of amorphous carbon (AC). Assuming thermal emission to be the dominant flux Figure 2 shows the best fit of Equation 1 obtained with component, the mid-IR photometry probes the peak of the IDL MPFIT function (Markwardt 2009), which mini- the blackbody produced by warm grains. The flux can be 2 mizes the value of χ by varying Md and Td. With only fit as a function of the dust temperature, Td, and mass, four photometry points at each epoch, we limit our fits Md, to a single component (see Figure 2). Table 2 lists the M B (T )κ (a) F = d ν d ν , (1) best-fit parameters for AC grains of size a =0.1 µm. ν d2 SNe Type Ia Mid-IR Emission 3

Days (SN 2005gj) Table 2 100 200 300 400 500 600 a . µ IR Fitting Parameters ( = 0 1 m amorphous carbon) SN 2002ic 8.4 SN 2005gj SN Epoch (days) Md (M⊙) Td (K) Ld (L⊙) 8 × ) 8.2 2002ic 795 0.012 590 1.64 10 O • 2002ic 1179 0.016 495 8.70×107 2002ic 1370 0.014 481 6.34×107 8 2005gj 139 0.001 845 1.06×10 Log(L/L 8.0 2005gj 365 0.002 847 1.55×108 2005gj 510 0.006 725 2.34×108 7.8

3. ANALYSIS AND DISCUSSION 600 800 1000 1200 1400 Figure 3 plots the corresponding IR luminosity evolu- Days (SN 2002ic) tion for each object. SN 2002ic remains bright (Ld > Figure 3. The late-time mid-IR luminosity evolution of SNe 8 10 L⊙) for more than 2 yr post-discovery, but continues 2002ic and 2005gj. The luminosity is derived by summing over the modified blackbody fits given in Equation 1 and shown in Fig- to fade throughout the observations. By contrast, SN ure 2. SN 2005gj exhibits an unanticipated rise more than a year 2005gj brightens after 1 year post-discovery. While we post-discovery, suggesting renewed shock interaction between the only have mid-IR photometry available at these epochs, forward shock and a pre-existing circumstellar shell. we explore the constraints these data provide on the cir- cumstellar environment. ultraviolet, and/or X-ray flux given by 3.1. Possible Origins and Heating Mechanisms 64 Bν (Td)κ(ν)dν The source of the mid-IR emission is warm dust, but L = ρar2σT 4 R (2) opt/UV/X 3 d SN B (T )Q (ν)dν the origin and heating mechanism of the dust are less R ν SN abs clear. The dust may be either newly formed or pre- for a dust bulk (volume) density ρ and an effective SN existing, and either shock or radiatively heated; see blackbody temperature T , where B is the Planck Fox et al. (2010) for a full discussion. To discriminate SN ν blackbody function, Qabs is the dust absorption effi- between possible scenarios, we assume a spherically sym- ciency, and κ(ν) is the dust absorption coefficient. Figure metric, optically thin dust shell and calculate the black- 4 in Fox et al. (2013) shows that the blackbody radii of 4 1/2 body radius, rbb = [Ld/(4πσTd )] , which sets a min- SNe 2002ic and 2005gj [r ≈ (0.5–1.0) ×1017 cm] at 8 bb imum shell size. A luminosity Ld ≈ 10 L⊙ and dust temperatures Td ≈ 500–750 K require optical and/or X- temperature Td ≈ 500 K yield a blackbody radius of 8 9 17 ray luminosities in the range 10 . Lopt/UV/X . 10 L⊙. rbb ≈ 10 cm. This radius is larger than that of a −1 While we do not have optical or X-ray observations at forward shock moving at vs ≈ 10,000 km s on day these epochs, the most recent optical observations of SNe 800. This point seemingly rules out, particularly for 2002ic and 2005gj are consistent, with measured lumi- 9.1 9.4 the younger SN 2005gj, the possibility that the major- nosities of 10 and 10 L⊙ on days 250 and 149, re- ity of the observed dust condensed in either the more spectively (Deng et al. 2004; Prieto et al. 2007). slowly moving ejecta or the cold, dense shell that can form behind the forward shock. [We note, however, 3.2. Evidence for Shells that Silverman et al. (2013b) observed evidence of some newly formed dust in SN 2005gj via increasing absorp- Mid-IR wavelengths probe the characteristics of the tion in the red Hα wing. While they do not estimate the CSM at the dust-shell radius. Assuming a dust-to-gas dust mass, the Spitzer observations suggest that this mass ratio expected in the H-rich envelope of a massive dust does not contribute significantly to the overall mid- star, Zd = Md/Mg ≈ 0.01, the dust-shell mass can be IR flux.] tied to the progenitor’s total mass-loss rate, The alternative to newly formed dust is a pre-existing Md shell. Again, the fact that the blackbody radius is be- M˙ outer = vw yond the forward-shock radius rules out the likelihood of Zd∆r shock heating. Furthermore, an IR light-echo scenario 3 Md vw = − (e.g., Dwek 1983), in which the dust shell is heated by 4M⊙ 120 km s 1  the peak SN luminosity, is unfeasible. The implied shell 16 5 × 10 cm r −1 radii (recho = tplateau/2c) would require peak luminosi- ×   M⊙ yr , (3) 11 r ∆r ties of nearly 10 L⊙ to heat the dust to the observed temperatures. for a progenitor wind speed vw. The relatively narrow In fact, the observed dust-shell parameters (i.e., ra- lines observed in SNe Ia-CSM originate in the slow pre- dius, temperature, mass) are comparable to those seen shocked CSM and can be used to approximate the pro- in SNe IIn (e.g., Fox et al. 2011, 2013). In these cases, genitor wind speed. The precursor wind velocities for −1 ongoing interaction between the forward shock and dense SNe 2002ic and 2005gj are vw = 100 and 60 km s , re- CSM produce optical, UV, and X-ray emission that ra- spectively (Kotak et al. 2004; Aldering et al. 2006). As- diatively heat a pre-existing dust shell. Assuming an suming a thin shell, ∆r/r = 1/10, wind speed vw = −1 17 optically thin dust shell, the observed dust temperature 60 km s , and radius rbb = 10 cm, the approxi- (Td) and shell radius (rd) require a combined optical, mate mass-loss rate to produce the observed dust shell 4

−2 −1 is M˙ outer ≈ 10 M⊙ yr . A smaller dust-shell radius While we observe a luminosity increase of this magni- would require an even larger mass-loss rate. tude (see Table 2), the inferred dust temperature actually The optical and/or X-ray emission generated by CSM decreases by ∼ 120 K. The implication is that the black- interaction traces the mass loss at the inner radii. As- body radius must also increase from 5 × 1016 to 1017 cm, suming a density ∝ r−2 wind profile, the rate can which would explain a lower dust temperature along with be written as a function of the optical/X-ray luminos- a higher dust mass and luminosity. The problem with ity, progenitor wind speed, and shock velocity (e.g., invoking this scenario is that the only way to increase Chugai & Danziger 1994; Smith et al. 2009): the blackbody radius (assuming a spherically symmetric shell of dust) would be for the increased optical luminos- ˙ 2vw ity to vaporize all dust out to the new blackbody radius. Minner = 3 Lopt/UV/X, ǫvs For a dust vaporization temperature of Tevap ≈ 2000 K, −4 Lopt/UV/X Equation 2 and Figure 8 of Fox et al. (2010) show that =2.1 × 10 17  41 −1  to vaporize dust out to 10 cm requires a luminosity of 3 × 10 erg s 10 −1 > 10 L⊙, which is not likely. The more probable expla- ǫ vw nation is that with only four data points, the slopes of × − 0.5 120 km s 1  the curves in Figure 2 and, therefore the temperatures, −3 vs −1 are biased by contamination from the underlying galaxy × M⊙ yr , (4) ′′ ′′ 104 km s−1  nucleus (< 1 away and a 0.6 pixel scale). The total dust luminosity, which is less sensitive to the slope, is where ǫ < 1 is the efficiency of converting shock kinetic most consistent with radiative heating by renewed shock energy into visual light. We assume a value ǫ ≈ 0.5, interaction. although the conversion efficiency can vary with wind density and shock speed. Again, we do not have late-time 4. SUMMARY optical or X-ray observations, but we do have theoretical estimates from Equation 2 in §3.1. An optical luminosity This Letter presents unpublished archival Spitzer data 8.5 −1 on SNe Ia-CSM 2002ic and 2005gj obtained > 1300 Lopt/UV/X ≈ 10 L⊙, wind speed vw = 120 km s , −1 and 500 days post-discovery. The mid-IR data show shock velocity vs = 10,000 km s (Deng et al. 2004), evidence of emission from warm dust. While we do not and conversion efficiency ǫ = 0.1 correspond to a mass- have simultaneous observations at shorter wavelengths, ˙ −3 −1 loss rate Minner ≈ 10 M⊙ yr . the warm-dust parameters are most constant with a While the variables used above are only order-of- pre-existing dust shell heated by a combination of ˙ magnitude approximations, the derived rate for Minner optical, UV, and X-ray emission continuously generated reveals two things about the circumstellar medium. (1) by ongoing CSM interaction. The degree of CSM The difference between the mass-loss rates (M˙ inner < interaction dictates the dust temperature and, thereby, the luminosity. In the case of SN 2005gj, the mid-IR lu- M˙ outer) suggests that the dust shells were formed dur- ing a period of increased, nonsteady mass loss. (2) Com- minosity nearly doubles more than 1 year post-discovery, pared to mass-loss rates derived from optical data at ear- suggesting an increasing amount of CSM interaction. −3 −1 We attribute this renewed shock interaction to a dense lier epochs, our estimate of M˙ inner ≈ 10 M⊙ yr is consistent with SN 2002ic on day 250 (for ǫ ≈ 0.1; circumstellar shell produced during a period of increased Kotak et al. 2004), but at least an order of magnitude mass loss by the progenitor companion. While progeni- larger than that measured for SN 2005gj on day 74 tor mass loss suggests a single-degenerate channel, such (Prieto et al. 2007). The increased circumstellar density rates have also been derived for both the core-degenerate derived for SN 2005gj at this late time suggests the pres- and double-degenerate models (Livio & Riess 2003; ence of another shell of material. Ilkov & Soker 2012, 2013; Soker et al. 2013; Shen et al. The decline of SN 2002ic occurs at >800 days, corre- 2013). Future multi-wavelength observations of SNe sponding to the time at which the forward shock (v = Ia-CSM will be necessary to better trace the complete s mass-loss history and constrain the nature of the com- 10,000 km s−1) would reach the blackbody radius (r ≈ bb panion star. 1017 cm). The declining light curve may be attributed to the forward shock overtaking and destroying the dust This work is based on archival data obtained with and/or a decreasing amount of CSM interaction accom- the Spitzer Space Telescope, which is operated by the panied by a declining radiative heating source. Jet Propulsion Laboratory, California Institute of Tech- Alternatively, the rebrightening of the pre-existing nology, under a contract with NASA. Support for this dust in SN 2005gj is likely due to radiative heating by work was provided by NASA through an award issued renewed shock interaction with this dense circumstellar by JPL/Caltech (P90031). We also acknowledge gener- shell. From Equation 4, an order-of-magnitude increase ous financial assistance from the Christopher R. Redlich in mass loss results in an increase in the optical lumi- Fund, the TABASGO Foundation, and NSF grant AST- nosity by a factor of 3–4, assuming the shock velocity 1211916. decreases to 0.8 of its former value. For a constant black- 16 body radius of rbb ≈ 5×10 cm, Equation 2 and Figure REFERENCES 10 in Fox et al. (2010) show that an increase in the opti- 8 8.4 cal luminosity from 10 to 10 L⊙ results in a dust tem- 4 Aldering, G., et al. 2006, ApJ, 650, 510 perature increase from ∼ 600 to 750 K. Since Ld ∝ Td , Blondin, S., Prieto, J. L., Patat, F., Challis, P., Hicken, M., this change in temperature results in a luminosity in- Kirshner, R. P., Matheson, T., & Modjaz, M. 2009, ApJ, 693, crease of a factor of 2.4. 207 SNe Type Ia Mid-IR Emission 5

Bloom, J. S., et al. 2012, ApJL, 744, L17 Kotak, R., Meikle, W. P. S., Adamson, A., & Leggett, S. K. 2004, Brown, P. J., Dawson, K. S., Harris, D. W., Olmstead, M., Milne, MNRAS, 354, L13 P., & Roming, P. W. A. 2012, ApJ, 749, 18 Livo, M. & Riess, A. G., 2003, ApJ, 594, 93 Chugai, N. N., & Danziger, I. J. 1994, MNRAS, 268, 173 Markwardt, C. B. 2009, in Astronomical Data Analysis Software Deng, J., et al. 2004, ApJ, 605, L37 and Systems XVIII, ed. D. A. Bohlender, D. Durand, & Dilday, B., et al. 2012, Science, 337, 942 P. Dowler (San Francisco: ASP), 251 Dwek, E. 1983, ApJ, 274, 175 Nugent, P. E., et al. 2011, Nature, 480, 344 Filippenko, A. V. 1997, ARA&A, 35, 309 Patat, F., et al. 2007, Science, 317, 924 Foley, R. J., et al. 2012a, ApJ, 752, 101 Phillips, M. M. 1993, ApJ, 413, L105 —. 2012b, ApJ, 744, 38 Prieto, J. L., et al. 2007, arXiv:0706.4088 Fox, O. D., Chevalier, R. A., Dwek, E., Skrutskie, M. F., Schlegel, E. M. 1990, MNRAS, 244, 269 Sugerman, B. E. K., & Leisenring, J. M. 2010, ApJ, 725, 1768 Shen, K., et al. 2013, arXiv, 1302.2916 Fox, O. D., Filippenko, A. V., Skrutskie, M. F., Silverman, J.M., Silverman, J. M., et al. 2013a, ApJ, submitted (arXiv:1303.7234) Ganeshalingam, M., Cenko, S. B., & Clubb, K. I. 2013, —. 2013b, ApJS, in press (arXiv:1304:0763) arXiv:1304.0248 Simon, J. D., et al. 2009, ApJ, 702, 1157 Fox, O. D., et al. 2011, ApJ, 741, 7 Smith, N., et al. 2009, ApJ, 695, 1334 Ganeshalingam, M., Li, W., & Filippenko, A. V. 2011, MNRAS, Soker, N., et al. 2013, MNRAS, 431, 1541 416, 2607 Wang, L., Baade, D., H¨oflich, P., Wheeler, J. C., Kawabata, K., Gerardy, C. L., et al. 2002, ApJ, 575, 1007 & Nomoto, K. 2004, ApJ, 604, L53 Hamuy, M., et al. 2003, Nature, 424, 651 Wood-Vasey, W. M., Wang, L., & Aldering, G. 2004, ApJ, 616, Hildebrand, R. H. 1983, QJRAS, 24, 267 339 Ilkov, M. & Soker, N. 2012, MNRAS, 419, 1695 Ilkov, M. & Soker, N. 2013, MNRAS, 428, 579