An Intermediate Type Ia Supernova Between Normal And Super-Chandrasekhar

3.7 MB·

by  Yi Cao

Draft version May 18, 2016
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
ABSENCE OF FAST-MOVING IRON IN AN INTERMEDIATE TYPE Ia SUPERNOVA BETWEEN NORMAL
AND SUPER-CHANDRASEKHAR
Yi Cao1, J. Johansson2, Peter E. Nugent3,4, A. Goobar5, Jakob Nordin6, S. R. Kulkarni1, S. Bradley Cenko7,8,
Ori D. Fox4,9, Mansi M. Kasliwal1,10, C. Fremling11, R. Amanullah5, E. Y. Hsiao12,13, D. A. Perley1,
Brian D. Bue14, Frank J. Masci15, William H. Lee16, Nicolas Chotard17
Draft version May 18, 2016
ABSTRACT
6 Inthispaper,wereportobservationsofapeculiarTypeIasupernovaiPTF13asv(a.k.a.,SN2013cv)
1 from the onset of the explosion to months after its peak. The early-phase spectra of iPTF13asv show
0
absence of iron absorption, indicating that synthesized iron elements are confined to low-velocity
2
regions of the ejecta, which, in turn, implies a stratified ejecta structure along the line of sight. Our

y analysisofiPTF13asv’slightcurvesandspectrashowsthatitisanintermediatecasebetweennormal
a andsuper-Chandrasekharevents. Ontheonehand,itslightcurveshape(B-band∆m =1.03±0.01)
15
M andoverallspectralfeaturesresemblethoseofnormalTypeIasupernovae. Ontheotherhand,similar
to super-Chandrasekhar events, it shows large peak optical and UV luminosity (M = −19.84mag,
B
7 M =−15.5mag)arelativelylowbutalmostconstantSiIIvelocitiesofabout10,000kms−1, and
uvm2
1 persistent carbon absorption in the spectra. We estimate a 56Ni mass of 0.81+0.10M and a total
−0.18 (cid:12)
ejecta mass of 1.59+0.45M . The large ejecta mass of iPTF13asv and its stratified ejecta structure
] −0.12 (cid:12)
R together seemingly favor a double-degenerate origin.
S Subject headings: supernovae: general – supernovae: individual (iPTF13asv, SN2013cv) – ultraviolet:
. general
h
p
- 1. INTRODUCTION established empirical relation between variation of their
o
peak magnitudes and light curve shapes (Phillips 1993),
r Type Ia supernovae (SNe) are thermonuclear explo-
t they are standardized to measure cosmological distances
s sions of carbon-oxygen white dwarfs (WDs). Since the
a majorityofthem(thenormalTypeSNeIa)followawell- (see Goobar & Leibundgut 2011 for a review). However,
[ the underlying progenitor systems and explosion mecha-
nisms of SNe Ia remain poorly understood.
2 1Astronomy Department, California Institute of Technology, Recent observations have provided mounting evidence
v Pasadena,CA91125,USA thatSNeIahavemultipleprogenitorchannels(seeMaoz
6 2Benoziyo Center for Astrophysics, Weizmann Institute of
et al. 2014 for a review). In the single-degenerate (SD)
8 Science,76100Rehovot,Israel
3ComputationalCosmologyCenter,ComputationalResearch channel, a WD accretes material from a companion
6
Division, Lawrence Berkeley National Laboratory, 1 Cyclotron star and explodes when its mass approaches the Chan-
0 Road,MS50B-4206,Berkeley,CA94720,USA drasekhar limit (Whelan & Iben 1973). This channel
0 4DepartmentofAstronomy,UniversityofCaliforniaBerkeley,
is supported by possible detections of companion stars
. Berkeley,CA94720-3411,USA
1 5Oskar Klein Centre, Physics Department, Stockholm Uni- in pre- or post-SN images (McCully et al. 2014; Foley
0 versity,SE-10691Stockholm,Sweden et al. 2014), likely signatures of SN-companion collisions
6 6Institut fu¨r Physik, Humboldt-Universit¨at zu Berlin, New- (Caoetal.2015;Marionetal.2016),andobservationsof
1 tonstr. 15,12489Berlin,Germany
7AstrophysicsScienceDivision,NASAGoddardSpaceFlight variable Na ID absorption (Patat et al. 2007; Sternberg
:
v Center,MailCode661,Greenbelt,Maryland20771,USA et al. 2014). In the double-degenerate (DD) channel, in
i 8Joint Space-Science Institute, University of Maryland, Col- contrast, two WDs collide or merge in a binary or even
X legePark,MD20742,USA
triple system to produce a SN Ia (e.g., Nomoto & Iben
9Space Telescope Science Institute, 3700 San Martin Drive,
r Baltimore,MD21218,USA 1985; Kushnir et al. 2013). This channel is consistent
a 10Observatories of the Carnegie Institution for Science, 813 with observations of two nearby Type Ia SN2011fe and
SantaBarbaraStreet,Pasadena,California91101,USA SN2014J(e.g.,Lietal.2011;Brownetal.2012;Shappee
11DepartmentofAstronomy,TheOskarKleinCenter,Stock-
etal.2013;Marguttietal.2014;Kellyetal.2014;Goobar
holmUniversity,AlbaNova,10691Stockholm,Sweden
12DepartmentofPhysics,FloridaStateUniversity,Tallahas- et al. 2015; Lundqvist et al. 2015). Despite these inter-
see,FL32306,USA estingconstraintsfromindividualevents,theprogenitors
13DepartmentofPhysicsandAstronomy,AarhusUniversity, of most SNe Ia are still unknown.
NyMunkegade120,8000AarhusC,Denmark
14JetPropulsionLaboratory,CaliforniaInstituteofTechnol- In the SD channel, rigid rotation may provide addi-
ogy,Pasadena,CA91125,USA tional support for a WD of a mass slightly larger than
15Infrared Processing and Analysis Center, California Insti- the Chandrasekhar limit and differential rotation may
tuteofTechnology,MS100-22,Pasadena,CA91125,USA support for an even more massive WD. However, the
16Instituto de Astronom´ıa, Universidad Nacional Aut´onoma
theoreticalviabilityofmassive, rotation-supportedWDs
de M´exico, Apdo. Postal 70-264 Cd. Universitaria, M´exico DF
04510,M´exico is much less clear in reality (Yoon & Langer 2004; Saio
17Universit´e de Lyon, F-69622, France ; Universit´e de Lyon & Nomoto 2004; Piro 2008; Justham 2011; Di Stefano
1, Villeurbanne ; CNRS/IN2P3, Institut de Physique Nucl´eaire
et al. 2011; Hachisu et al. 2012). In the DD channel, in
deLyon
2 Cao et al.
contrast, the exploding WD binary may allow SN ejecta
massmuchhigherthantheChandrasekharlimit. Infact, Table 1
PhotometryofiPTF13asv
more than a handful of SNe were found to have total
ejecta masses significantly exceeding the Chandrasekhar
Tel./Inst.1 Filter MJD−56000 mag.2 mag. err.
limit (Howell et al. 2006; Hicken et al. 2007; Yuan et al.
(day)
2010; Yamanaka et al. 2009; Scalzo et al. 2010; Silver-
man et al. 2011; Scalzo et al. 2014c). However, these P48 PTF-R 412.456 >20.35 ···
P48 PTF-R 413.442 20.44 0.17
super-Chandrasekhar SNe show distinctive characteris-
P48 PTF-R 413.471 20.34 0.16
tics compared to normal events: they are overluminous P48 PTF-R 414.470 19.58 0.09
in both the optical and UV, implying a large amount P48 PTF-R 415.469 19.06 0.06
of synthesized 56Ni. They show low expansion velocities ···
and long rise times, leading to massive ejecta. They also 1 This column lists the telescopes and instruments used for
show persistent absorption from unburned carbon. photometric observations of iPTF13asv. We built reference
images by stacking pre-SN or post-SN frames and used an
In this paper, we present observations of a peculiar
image subtraction technique to remove light contamination
SN Ia, iPTF13asv, which shares observational charac- fromthehostgalaxy. Point-spreadfunction(PSF)photom-
teristics with both super-Chandrasekhar and normal SN etryisperformedonsubtractedimages. Thephotometryis
Ia. It was discovered with r = 20.54 ± 0.16mag at calibratedeithertoSDSSorbyobservingLandoltphotomet-
ricstandardstars.
α = 16h22m43s.19, δ = +18◦57(cid:48)35(cid:48)(cid:48).0 (J2000) in the 2 Conventionally, the magnitudes in UBVJH bands are in
vicinity of galaxy SDSSJ162243.02+185733.8 on UTC theVegasystem. ThoseinotherbandsareintheABsystem.
2013 May 1.44 (hereafter May 1.44) by the intermedi- Noextinctioniscorrectedinthiscolumn.
ate Palomar Transient Factory (iPTF; Law et al. 2009;
Rau et al. 2009). Nothing was seen at the same loca-
tion down to 5-σ detection thresholds of r (cid:39) 21.0mag
on images taken on April 30.5 and earlier. iPTF13asv SSSSSSSSSSS SSSS SS S
was independently discovered and classified as a pecu-
10
liar type Ia by Zhou et al. (2013), and was designated as P48 upper limit RATIR i+0.7
11 P48 R RATIR r
SN2013cv. P60 B 3.5 RATIR z+2.0
−
This paper is organized as follows: the observational 12 P60 g−2.5 RATIR J+5.0
P60 r RATIR H+7.0
data are presented in §2. The photometric and spectro- 13 P60 i+0.7 RATIR Y+3.0
scopicpropertiesareanalyzedin§3,andweconstructits 14 NNOOTT UB−34 SSwwiifftt uuvvmw22−1101
biinnol§§o54m.aenAtdridcoiuslicrguhcstosincoucnlruvosefioatnnhsdeaenrsaettisumurametmeoftahirPeizTteodFt1ain3lae§sj6ve.citsagmivaesns + Offset1156 NOT V−−1 −
In order to have a comparison to other SNe, de 17
u
we adopt a fiducial value of the Hubble constant nit18
g
H0 = 72kms−1Mpc−1. The apparent host galaxy of Ma19
iPTF13asv does not have a redshift-independent dis- nt
a20
tance measurement in the NASA/IPAC Extragalactic er
p
Database. Thus the redshift 0.036 leads to a distance Ap21
modulus of 35.94mag. The peculiar motion of the 22 200
hofos(cid:47)t g0a.0la5xmyaagtt∼o1t0h0ekdmisst−a1ncientmroodduucleuss.anThuencGeratlaaicnttiyc 23 Jy)150
24 µ100
line-of-sight extinction is E(B−V)=0.045 (Schlafly & 25 f(ν 50
Finkbeiner2011). WecorrectfortheGalacticextinction
0
by using the parameterized model in Fitzpatrick (1999) 0 1 2 3 4 5 6 7
MJD - 56408 (days)
with R =3.1.
V 20 0 20 40 60 80 100 120 140
MJD 56430.2 (days)
2. OBSERVATIONS −
The nightly cadence survey of iPTF (weather permit-
Figure 1. Multi-colorlightcurveofiPTF13asvColorsandshapes
ting)withthe48-inchtelescopeatthePalomarObserva-
represent different filters and instruments, respectively. A devia-
tories (P48) provides a well-sampled R-band light curve tion of (cid:39) 0.1mag between the P48 R-band (red circles) and P60
of iPTF13asv covering the pre-SN history and its rise r-band (red diamonds) is due to the difference between the P48
phase. After discovery, we also utilized the Palomar 60- MouldRfilterandtheP60SDSSrfilter. The“S”ticksonthetop
axisdenotespectroscopicobservationepochs. Thesolidcurvesare
inch telescope (P60; Cenko et al. 2006), the Andalusia
SALT2 best-fit light curves in corresponding filters. The dashed-
Faint Object Spectrograph and Camera (ALFOSC) on dotted curves are the IR template from Stanishev et al. (2015).
the Nordic Optical Telescope (NOT), and the RATIR The inset zooms into the very early phases of the PTF R-band
light curve. The dashed curve in the inset shows the best t2 law
camera mounted on the OAN/SPM 1.5-meter Harold L.
fittotheearlylightcurve.
Johnson telescope for multi-band photometric follow-up
observations. We also triggered target-of-opportunity
observations of the Swift spacecraft for X-ray and UV SN Integral Field Spectrograph (SNIFS; Lantz et al.
follow-up. Theground-basedphotometricmeasurements 2004) on the 2.2m telescope of the University of Hawaii,
are presented in Table 1 and the space-based measure- theDualImagingSpectrograph(DIS)ontheARC3.5m
ments in Table 2. telescopeatApachePointObservatory(APO),theDou-
Spectroscopic observations were undertaken with the bleSpectrograph(DBSP;Oke&Gunn1982)onthe200-
iPTF13asv 3
Table 2
SwiftObservations
Obs. Date UVOT/uvm21 UVOT/uvw21 XRT2
exp. time(s) mag(AB) exp. time(s) mag(AB) exp. time(s) counts(cntss−1)
May17.4 1386 21.02±0.08 1540 20.59±0.08 2971 <3.6×10−3
May22.6 1154 21.25±0.10 1193 20.92±0.09 2382 <8.0×10−3
May25.9 1298 21.83±0.12 1274 21.39±0.10 2625 <4.3×10−3
June02.7 646 <22.20 604 <22.36 1264 <9.0×10−3
June10.5 462 <21.98 464 <22.18 928 <1.2×10−2
June13.4 612 <22.37 674 <22.46 1326 <8.7×10−3
1 We used the HEASoft package to perform aperture photometry on the UVOT images. The photometry is
correctedforcoincidentlossandwiththePSFgrowthcurveandcalibratedwiththelatestcalibration(Breeveld
et al. 2011). In order to remove host galaxy contamination in the photometric measurements, we acquired
post-SNreferenceframes. Incasesofnon-detection,weestimated3-σ upperlimits.
2 WeusedtheXImagesoftwaretoanalyzetheXRTdata. Incasesofnon-detection,weestimateupperlimits
witha99.7%confidencelevel.
inch Hale telescope (P200) at Palomar Observatory, and (or equivalently, −13.7 days with respect to the B-band
theFolded-portInfraRedEchellette(FIRE)ontheMag- maximum which is determined in §3.3). As highlighted
ellan Baade Telescope at Las Campanas Observatory. in gray in Figure 3, the spectrum shows no signature of
The spectral sequence is presented in Figure 2. either FeII or FeIII.
The light curves and spectra are made publicly avail- We also compare early-phase spectra of iPTF13asv
able via WISeREP18 (Yaron & Gal-Yam 2012). to those of well-studied normal and overluminous Type
Ia events in the top panels of Figure 4. Most spec-
3. ANALYSIS
tra in comparison, including the spectrum of the super-
3.1. Initial Rise and Explosion Date Chandrasekhar SN2009dc at −7 days, clearly show the
InordertodeterminetheexplosiondateofiPTF13asv, existence of iron absorption features. The exceptions
we follow Nugent et al. (2011) and model the early PTF are two super-Chandrasekhar events, SN2006gz and
R-band light curve of iPTF13asv as a freely expanding SN2007if, which have weak or no iron absorption.
fireball where the luminosity increases as ∝ t2 and the As a result of nucleosynthesis and mixing during SN
temperature remains constant. Restricting ourselves to explosions, iron commonly manifests itself as absorption
the light curve within four days of discovery, we find features in SN spectra, either as Fe II at low effective
a best fit (the inset in Figure 1) at an explosion date temperatures or as Fe III at higher temperatures. The
of April 29.4±0.3 (95% confidence interval) with a fit- absence of iron at early phases implies that weak mixing
ting χ2 = 4.3 for five degrees of freedom. The best-fit during the SN explosion confines synthesized iron group
elements in low-velocity regions of the ejecta.
light curve is also consistent with the non-detection up-
The centric concentration of iron can be verified by
per limit on April 30.4.
If we generalize the t2 model with a power-law model, strong UV emission at the same time, as the iron
group elements are the main absorbers of photons below
we obtain strong degeneracy between the explosion date
andthepower-lawindexoveralargerangeoftheparam- 3500˚A.However,wedidnottriggerSwift observationsat
eter space. In fact, Firth et al. (2015) analyzed the rise early phases because the SN is located beyond our trig-
behavior of a large sample of SNe Ia with the power- ger criteria of 100Mpc. In comparison to the spectral
law model and also found that the power-law indices shape of SN2011fe, the spectral shape of iPTF13asv at
have large uncertainties with a mean value of 2.5. This −7 days (top right panel of Figure 4) indicates stronger
is probably because shallow-deposited 56Ni heats up SN fluxes at shorter wavelengths, hinting a strong emission
photospheres. Furthermore, there could be a dark time in the UV.
between the SN explosion and the SN light curve pow- TheweakmixingofironintheSNexplosionmayhave
ered by radioactive decay on the diffusive timescale for strongimplicationsfortheexplosionmechanismandwill
theshallowestdepositionof56Niintheejecta(Piro2012; be discussed in §5.3. As shown in the bottom two pan-
Piro & Morozova 2015). Hence, it is nontrivial to esti- els of Figure 4, iron features appear in the iPTF13asv
matetheexactexplosiondatepurelyfromtheearlylight spectra around and after maximum. In the next few
curve. Sincethefollowinganalysisanddiscussionarenot subsections, we investigate the specifics of iPTF13asv.
very sensitive to the exact explosion date, for simplicity,
weadopttheexplosiondatedeterminedbythet2 model. 3.3. Light Curves
3.2. Absence of Iron in Early-phase Spectra Inordertodeterminethelightcurveshapeparameters
of iPTF13asv, we use the SALT2 software (Guy et al.
The most striking feature of these early-phase spec-
2007) to fit its optical light curve (see Figure 1 for the
tra is the absence of iron absorption. We used SYN++
SALT2 best-fit light curves). The best-fit light curve
(Thomas et al. 2011a) to perform spectral feature iden-
givesarest-frameB-bandpeakmagnitudem =16.28±
tification on the spectrum taken 11 days after explosion B
0.03 on May 18.12±0.09. We set this B-band peak date
18 WISeREP is available at http://www.weizmann.ac.il/ as t = 0 in the rest of this paper. The fit also gives a
astrophysics/wiserep/. color term c = −0.16±0.02, and two shape parameters
4 Cao et al.
iPTF13asv C II
S II Fe III
Si II Fe II
Mg II Ca II
Si III
et
Offs
-13.7 SNIFS +
-8.7 SNIFS d Fλ
e
-6.7 SNIFS al
c
S
-3.8 SNIFS
-1.7 SNIFS
0.0 APO
3.3 SNIFS
3500 4000 4500 5000 5500 6000 6500 7000
6.3 SNIFS Rest-Frame Wavelength ( )
t
e 12.0 APO Figure 3. SYN++syntheticSNspectrumfittoiPTF13asvat-9.2
s
+off 13.3 SNIFS dspaeycst.rFumromwtitohpotuotbFoett(osmoliadr)e,ththeeobsysenrtvheedtiscpespctercutmru,mthwesityhntFheetiIcI
λ 16.1 P200 (dashed),andthesyntheticspectrumwithFeIII(dashed-dotted).
gf The absorption features from different species are illustrated in
lo 18.3 SNIFS differentcolors. TheFeIIandFeIIIabsorptionwavelengthranges
33.3 SNIFS arehighlightedinlightgray.
38.2 SNIFS
40.0 APO 1.03mag) and SN2011fe (∆m15 = 1.10mag). Since
SN2011fe is unreddened by its host (Nugent et al. 2011;
43.2 SNIFS
Vink´o et al. 2012), similar colors around maximum sug-
46.1 SNIFS
gestthatiPTF13asvalsohaslittlelocalextinction. After
51.1 SNIFS the maximum, iPTF13asv has a slightly blue color com-
58.2 SNIFS pared to SN2011fe, probably due to the different stretch
of these two events (Nobili & Goobar 2008), until they
join the Lira relation after +30 days. Second, the intrin-
FeII/III
Ca II FeII/IIIS II Si II Ca II sic color B−V =0.95mag of iPTF13asv at +35 days is
3000 4000 5000 6000 7000 8000 9000 10000 11000 consistent with the latest calibration of the Lira relation
Rest-frame wavelength ( ) (Burns et al. 2014). Third, the absence of Na ID ab-
sorptioninthelow-resolutionopticalspectraalsoimplies
weakextinctioninthehostgalaxy. Givenatypicalveloc-
28.0 FIRE ity dispersion of 10kms−1 for a dwarf galaxy (see §3.5
for a discussion of the iPTF13asv host galaxy; Walker
et al. 2007), we derive from the highest signal-to-noise
ratio spectrum that the equivalent width for each of the
0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 NaI D lines is less than 0.2˚A(5-σ). Using the empirical
Rest-Frame Wavelength (µm)
relation in Poznanski et al. (2012), we find that the ex-
tinction E(B −V) < 0.06. Therefore, in what follows,
we neglect the local extinction correction.
After correction for Galactic extinction we derive an
Figure 2. Optical and near-IR spectral evolution of iPTF13asv. absolute peak magnitude of iPTF13asv in its rest-frame
Ticksatthebottomdenotethemainspectralfeatures. Thephases
B-band to be −19.84 ± 0.06. This is about 0.5mag
andtelescopes/instrumentsarelabeledtotherightofcorrespond-
ingspectra. Thelong-slitspectratakenbyAPO/DIS,P200/DBSP brighter than normal SNe Ia at peak.
andMagellan/FIREareextractedthroughtheusualproceduresin The k-correction in the UV and optical wavelengths is
IRAF and/or IDL and calibrated using observations of spectro- negligible. Synthetic photometry using both the Nugent
scopicfluxstandardstars. DatareductionofSNIFSisoutlinedin
SN Ia template (Nugent et al. 2002) and the observed
Alderingetal.(2006). However,duetobadweather,fluxcalibra-
tionofSNIFSspectrawasnotcomplete. Therefore,weinterpolate HST UV spectra of SN2011fe (Mazzali et al. 2014).
themulti-bandlightcurvesand“warp”thespectrawithlow-order shows that the k-correction is less than 0.1mag in the
polynomialstomatchphotometricdata. optical and 0.2mag in the Swift/UVOT UV filters.
In Figure 6, the optical light curves of iPTF13asv are
x =0.0055±0.0001andx =0.37±0.09. Basedonthe compared to those of well-studied SNe, including nor-
0 1
fitted SALT2 light curve, we derive a color (B−V) = mal SN2011fe; overluminous SN1999aa and SN1991T;
0
−0.14±0.03 at the B-band maximum. We also obtain and super-Chandrasekhar SN2006gz, SN2007if, and
∆m =1.03±0.01fromx byusingtherelationinGuy SN2009dc. All the light curves have been offset to
15 1
et al. (2007). match their peak magnitudes and to the epoch of the
The local extinction in the host galaxy of iPTF13asv B-band maxima. Figure 6 illustrates that (1) the light
is probably minor for several reasons. First, Figure 5 curve width of iPTF13asv is similar to those of nor-
compares the B−V colors between iPTF13asv (∆m = maleventsandnarrowerthanthoseofoverluminousand
15
iPTF13asv 5
iPTF13asv
C II SN1991T C II
Si II SN1999aa
Ca II Fe II/Fe III SN2006gz Si II
Fe II/Fe III SN2011fe
SN2007if
SN2009dc
Ca II
Fe II/Fe III
onst −13d −7d Fe II/Fe III
c
+
fλ
g
o
l
Ca II Ca II
Fe II/Fe III Fe II/Fe III
Si II C II Si II C II
0d Fe II/Fe III 7d Fe II/Fe III
3000 4000 5000 6000 7000 8000 9000 3000 4000 5000 6000 7000 8000 9000 10000
Rest-frame wavelength (Å)
Figure 4. SpectralcomparisonofiPTF13asvtonormalSN2011fe(Pereiraetal.2013);overluminousSN1991T(Mazzalietal.1995)and
SN1999aa (Matheson et al. 2008); and super-Chandrasekhar events SN2006gz (Hicken et al. 2007), SN2007if (Blondin et al. 2012), and
SN2009dc(Taubenbergeretal.2011). Thephasesofthespectraareshownatthelowerleftcornerofeachpanel.
to the super-Chandrasekhar SN2006gz in the B- and V-
1.5 band light curves, but SN2006gz has a much stronger
near-IR secondary peak.
Figure 7 compares iPTF13asv in the Swift/UVOT
uvm2 and uvw2 filters to a large sample of both nor-
1.0
mal,overluminous,andsuper-ChandrasekharSNeIaob-
served by Swift (Milne et al. 2013; Brown 2014). While
theuvw2filterhasanon-negligibleleakageinlongwave-
V
B−0.5 lengths, the uvm2 filter does not have a significant leak-
age and therefore provides the best available measure-
ments of the UV flux. The figure shows that, like super-
0.0 Chandrasekhar events, iPTF13asv is more luminous in
the UV than the majority of normal events. Further-
SN2011fe more, Milne et al. (2013) divided SNe Ia into different
iPTF13asv
subclasses based on their Swift/UVOT colors. We can-
0.5
20 10 0 10 20 30 40
not make a direct comparison here because only uvm2
Phase (Day)
and uvw2 data are available for iPTF13asv. An indirect
Figure 5. B−VcolorevolutionofSN2011feandiPTF13asvafter
comparisonisthatiPTF13asvisbrighterthanSN2011fe
correctionforGalacticextinction. ThesolidlineshowstheLirare-
lationfromBurnsetal.(2014)andthedashedlinescorresponding by half a magnitude in the optical and by (cid:39) 0.7mag in
toits0.06magscattering. the UV. Since SN2011fe belongs to the NUV-blue sub-
class (Brown et al. 2012), iPTF13asv probably also be-
longs to the same subclass.
super-Chandrasekhar events, except for SN2006gz, (2)
We also compare the IR light curves of iPTF13asv in
iPTF13asvshowsanisolatedsecondarymaximuminthe
the J- and H-band to the most recent light curve tem-
I-band whose strength is weaker than those observed in
plate for normal Type Ia events (Figure 1; Stanishev
SN2011feandSN1991T,and(3)iPTF13asvmatcheswell
6 Cao et al.
1
B iPTF13asv V
SN2009dc
0 SN2011fe
SN1991T
SN2006gz
1 SN1999aa
SN2007if
2
x
a
m
3
m
−
m 4
≡ R I
ag 0
m
∆
1
2
3
4
20 0 20 40 60 80 20 0 20 40 60 80
Time Since B-band Maximum (days)
Figure 6. Light curve comparison of iPTF13asv to normal SN2011fe (Pereira et al. 2013); overluminous SN1991T (Nugent template)
andSN1999aa(Krisciunasetal.2000);andsuper-ChandrasekhareventsSN2006gz(Hickenetal.2007),SN2007if(Scalzoetal.2010),and
SN2009dc(Silvermanetal.2011). ThelegendcolorsarethesameasinFigure4.
et al. 2015) and find that the sparsely sampled light
curves of iPTF13asv roughly follow the template. The
18 uvm2
peak magnitudes of iPTF13asv is M = −18.83±0.08
J
16
and M =−18.16±0.19, compared to the median peak
H
14
magnitudesofM =−18.39andM =−18.36withrms
J H
12
scatters of σ =0.116±0.027 and σ =0.085±0.16 for
normal TypeJIa SNe (Barone-NugenHt et al. 2012). The ude 10
secondary maximum of iPTF13asv is clearly seen in J- gnit 8 SN200S9Nd2c01S1Nd2e011aa
band and H-band light curves as well as the I-band light e Ma 18 LSQ12iPgTdFj13SaNs2v012dn uvw2
curve,indicatingconcentrationofirongroupelementsin olut 16 SNM2il0n1e1 efte aSl.N (22001123c)g
the central region of the ejecta (Kasen 2006). This is in Abs 14
accordance with the absence of iron in the outer ejecta.
12
3.4. Spectra 10
3.4.1. Spectral Cross-matching 8
20 0 20 40 60 80
We use the latest versions of both SN Identification Days Since B-band Maximum
(SNID; Blondin & Tonry 2007) and SuperFit (Howell Figure 7. UV light curve comparison between iPTF13asv and
et al. 2005) to spectroscopically classify iPTF13asv. Be- other SNe Ia. A sample of SNe Ia from Milne et al. (2013)
is shown in gray. Highlighted by different colors are light
fore −5 days, not surprisingly, neither tools find a good
curves of iPTF13asv (black); SN2011fe (sky blue; Brown et al.
match for iPTF13asv spectra partly because they do 2012); SN2012cg (blue); super-Chandrasekhar events SN2009dc,
not have many early-phase SNe in their templates and SN2011aa, and LSQ12gdj (purple; Brown et al. 2014); and the
partlybecausetheironabsorptionisabsentintheearly- most UV-luminous event SN2011de (orange; Brown 2014). All
magnitudesareintheABsystem.
phase spectra of iPTF13asv. Around and after maxi-
mum, based on the first 5 best matches, both SNID and
SuperFit find that iPTF13asv spectroscopically resem-
iPTF13asv 7
bles normal SNe Ia (Table 3).
14000
3.4.2. Spectral comparison to well-studied SNe
13000
Figure 4 compares iPTF13asv spectra to those of
well-studied SNe at different epochs: normal SN2011fe s)12000
m/
(Pereiraetal.2013);overluminousSN1991T(Filippenko k11000
et al. 1992) and SN1999aa (Garavini et al. 2004; Math- y (
eson et al. 2008); and super-Chandrasekhar SN2006gz cit10000
o
(Hicken et al. 2007), SN2007if (Scalzo et al. 2010), and el 9000
V iPTF13asv
SN2009dc (Taubenberger et al. 2011). At −13 days Si 8000 normal
(top left panel of Figure 4), although both SN2006gz
and iPTF13asv have weak or no absorption from iron, 7000 super-Chandrasekhar
iPTF13asv shows strong Ca II H and K absorption but 6000
SN2006gz does not. In comparison, both SN1999aa and 17.5 18.0 18.5 19.0 19.5 20.0 20.5
B-band Peak Magnitude
SN2011fe at similar phases have both strong Ca II and
iron absorptions. In addition, the absorption of C II is Figure 8. SiIIvelocitiesatmaximumvs. peakmagnitudes. The
gray points are measurements taken from Foley et al. (2011) and
apparently weaker in iPTF13asv than in SN2006gz.
Scalzoetal.(2014b). NotethatFoleyetal.(2011)didnotcorrect
At a week before maximum (top right panel of Fig- the local extinction. The red points are taken from Scalzo et al.
ure 4), except for the prominent absence of iron in (2012).
iPTF13asv, the overall spectral features of iPTF13asv
are similar to those of normal SNe. Unlike the near-
absence of Ca II and Si II lines in SN1991T, iPTF13asv 250
shows apparent CaII and SiII absorption, the strengths ay)
of which are weaker than those seen SN1991T. Besides s/d 200
this, its CII feature becomes weaker. m/
k 150
finAdroguonoddmspaxecimtruaml m(laotwcehreslefbtetpwaeneenl oifPTFiFg1u3raesv4),awnde ent ( 100
SN2011fe. Atthisepoch,SN1991Tisalsobecomingsim- di
ilartonormalSNeIa. ThestrengthofSiIIabsorptionin Gra 50 iPTF13asv
iaPnTdFth13easstvroinsgboetnweeiennStNhe20w1e1afek.absorption in SN1991T ocity 0 normal
One week after maximum (lower right panel of Figure Vel 50 super-Chandrasekhar
4),thespectrumofiPTF13asvisverysimilartothoseof Si
100
normal events, but with strong CII absorption. 17.5 18.0 18.5 19.0 19.5 20.0 20.5
B-band Peak Magnitude
3.4.3. SiII velocities Figure 9. SiII velocity gradients at maximum vs. peak magni-
Wefurthermeasuretheexpansionvelocityevolutionof tude. The gray points are measurements taken from Foley et al.
(2011). Note that Foley et al. (2011) did not correct the local
iPTF13asv by fitting a Gaussian kernel to the SiII6355
extinction. TheredpointsaretakenfromScalzoetal.(2012).
line in each spectrum. The continuum is modeled by
a linear regression to regions at both sides of the line.
phases (Thomas et al. 2011b; Parrent et al. 2011; Silver-
Then we fit a linear model to the velocity measurements
man & Filippenko 2012). These C II features usually
between −10 and +10 days and estimate a velocity of
disappear before maximum. In contrast, some super-
(1.0±0.1)×104kms−1 and a velocity gradient close to
Chandrasekhar events show strong and persistent C II
zero at peak.
features even after maximum. Figure 11 compares the
Figure 8 shows that Si II velocities at maximum ver-
spectra of iPTF13asv at one week after maximum to
sus peak magnitudes. As can be seen in the figure,
those of well-studied SNe at similar phases. As can be
iPTF13asv has a Si II velocity lower than the majority
seen,neitherSN1991TnorSN1999aahasthecarbonfea-
of normal SNe Ia and similar to super-Chandrasekhar
ture at this phase; the carbon signature of iPTF13asv is
events. Figure 9 compares Si II velocity gradients at
not as strong as those seen in the super-Chandrasekhar
maximum versus peak magnitudes. Again, like super-
SN2009dc.
Chandrasekhar events, iPTF13asv has a velocity gra-
dient close to zero, lower than the majority of normal
3.5. Host Galaxy
events.
AfteriPTF13asvfadedaway,weobtainedalowsignal-
3.4.4. Carbon signatures to-noise ratio spectrum of its apparent host galaxy
We also note that iPTF13asv shows weak but persis- SDSSJ162254.02+185733.8. The spectrum only shows
tent C II absorption features until at least a week after Hα emission at the redshift of iPTF13asv. We fit a
maximum. Figure 10 shows SYN++ fits to iPTF13asv Gaussian profile to the Hα line and measure a lumi-
spectra of high signal-to-noise ratios, demonstrating the nosity of 3×1038ergss−1. We adopt the empirical re-
existence of CII6580 and CII7234 lines. The velocities lation between Hα luminosity and star formation rate
of these CII lines evolve from (cid:39)14,000kms−1 at −13.7 (Kennicutt 1998) and obtain a star formation rate of
days to (cid:39)11,000kms−1 at +6.3 days. 2×10−3M yr−1 for the host galaxy.
(cid:12)
About 30% of normal SNe are estimated to reveal Next, we construct the spectral energy distribution
the C II6580 and C II7234 absorption notches in early (SED) of the host galaxy with optical photometry from
8 Cao et al.
Table 3
SNIDresults
Phase FirstFiveBestMatches1
−6.8 05eu@−5.2(normal) 03ic@−4.1(normal) 08Z@−4.3(normal) 05na@−1.5(normal) 06cc@−9.7(normal)
0.0 96ai@+2.2(normal) 07F@+3.0(normal) [email protected](normal) 03cg@−2.1(normal) 94ae@+0.9(normal)
+6.8 08Z@+7.6(normal) 05na@+3.4(normal) 99aa@+1.8(peculiar) 01fe@+6.2(normal) 06cz@−2.0(91T-like)
+13.3 08Z@+12.3(normal) 07ca@+13.3(normal) 07bj@+12.0(normal) 03fa@+13.4(91T-like) 03kf@+13.4(normal)
1 Theformatinthiscolumnisname@phase(subclass).
2.51e−13
C II 6580 C II 7234
2.0
-13.7 d
)
-8.7 d 2cm1.5
ogf+offsetλ -6.7 d 1¸F(ergss¡¸1.0
l
3.3 d
0.5
6.3 d
1.5 1.0 0.5 0.0 1.5 1.0 0.5 0.0 0.0
4000 6000 8000 10000 12000 14000 16000 18000
Velocity (104kms−1) Velocity (104kms−1) Observed Wavelength (Å)
6000 6200 6400 6600 6800 7000 7200 7400
Figure 12. SED fit of the host galaxy. The data points (red)
Rest-Frame Wavelengths (Å)
are SDSS model magnitudes in optical and aperture-photometric
Figure 10. Carbon features of iPTF13asv at different phases. measurements in the near-IR RATIR reference images. The blue
The numbers to the right of each spectrum indicate the phases spectrumisthebestfitfromFAST.
in days. The observed spectra are shown in gray. The SYN++
spectra without C II are in blue and those with C II are in red.
The green dashed axes show velocities of C II6580 and C II7234 SDSS and near-IR photometry measured on the SN ref-
lines. erence images. The SED is then modeled with a galaxy
synthesiscodecalledtheFittingandAssessmentofSyn-
thetic Templates (Kriek et al. 2009) assuming an ex-
ponentially decaying star formation history and a so-
lar metallicity. The best-fit model gives a galaxy age
of 108.6years and a stellar mass log (M /M ) =
10 stellar (cid:12)
7.85+0.5 with a reduced χ2 =1.6 (Figure 12). The best-
−0.4
fit model also shows no ongoing star forming activity.
Because SED fitting models are usually insensitive to
very low star-forming rates, the best-fit model is con-
SN1999aa sistentwiththelowstarformationratederivedfromthe
Hα flux. The derived star formation rate and the stellar
mass of the iPTF13asv host galaxy follow the empiri-
t SN1991T
ns calrelationbetweenstellarmassandstarformationrate
o
c (Foster et al. 2012).
+
Since the host galaxy spectrum does not show
λ iPTF13asv
f [N II] lines, we estimate an upper limit of
g
o log([N II6548/Hα]) < −0.87. Using Denicol´o et al.
l SN2006gz
(2002), we derived a metallicity upper limit of 12 +
log(O/H)<8.3. Infact,usingthemass-metallicityrela-
tion (Foster et al. 2012), we estimate a gas-phase metal-
SN2007if
licityof12+log(O/H)∼8forthehostgalaxy. Compared
C II C II tothehostgalaxysamplesofSNeIainPanetal.(2014)
SN2009dc andWolfetal.(2016),SDSSJ162254.02+185733.8isone
of the least massive and most metal-poor galaxies that
6000 6200 6400 6600 6800 7000 7200 7400
host SNe Ia.
Rest-frame wavelength (Å)
4. BOLOMETRIC LIGHT CURVE AND EJECTA
Figure 11. Comparison of carbon features among iPTF13asv,
MASS
SN1991T, SN1999aa, SN2006gz, SN2007if, and SN2009dc at one
weekaftermaximum. 4.1. Bolometric Light curve
iPTF13asv 9
factor of order unity, depending on the distribution of
2.51e43 56Ni (Jeffery et al. 2006). We adopt a fiducial value of
α = 1.3 following Scalzo et al. (2012). The radioactive
power of 56Ni→56 Co→56 Fe is (Nadyozhin 1994)
ergs/s)2.0 S(tr)=[6.31exp(−tr/8.8)+1.43exp(−tr/111)]MNi ,
e ( (2)
curv1.5 where S(t) is in units of 1043ergs−1 and MNi is in units
ht of M . With the measured maximum bolometric lumi-
g (cid:12)
etric Li1.0 nesotsiimtyatLemaax56=Ni(2m.2as±s0o.f2)(0×.7170±430er.0g7s)−M1 at. −2.6days, we
m (cid:12)
olo About one month after the SN maximum, the SN de-
B0.5 brisexpandsapproximatelyinahomologousmanner. At
thistime, most56Niatomshavedecayedto56Co. Hence
the total luminosity can be approximated by
0.0
20 10 0 10 20 30 40 50 60
Phase (day) L(t)=(cid:2)1−exp(−(t /t)2)(cid:3)S (t)+S (t) , (3)
0 γ e+
Figure 13. Bolometric light curve. The measured bolometric
luminositiesatdifferentphasesareinblackcircles. Thebluecurve where S and S are the decay energy of 56Co carried
γ e+
is the best fit from the Gaussian process regression. The gray by γ-ray photons and positrons. At time t , the mean
regionrepresentsthe1-σ uncertaintyoftheregressioncurve. 0
opticalpathofγ-rayphotonsbecomesunity. Foragiven
density and velocity profile, t reflects the column den-
GiventhewavelengthcoverageoftheiPTF13asvspec- 0
sity along the line of sight. We fit equation (3) to the
tra, we first construct a pseudo-bolometric light curve
bolometric light curve of iPTF13asv after +20 days and
between3500and9700˚A.Inordertocalibratetheabso-
obtained t =44.2±2.0days.
0
lute fluxes of these spectra, we use interpolated optical
Next, we estimate the total ejecta mass of iPTF13asv.
lightcurvesto“warp”thespectra. Thenthespectraare
If we assume a density profile ρ(v) ∝ exp(−v/v ) where
e
integrated to derive the pseudo-bolometric light curve.
v is a scale velocity, then the ejecta mass can be ex-
e
DuetothesparselysampledUVandIRlightcurves,it
pressed as
isdifficulttoestimatetheUVandIRradiationatdiffer- 8π
ent phases. Therefore we calculate optical-to-bolometric Mej = κ q (vet0)2 , (4)
correction factors with a spectral template (Hsiao et al. γ
2007). In this calculation, we find that the UV correc- where κ is the Compton scattering opacity for γ-ray
γ
tionreachesabout25%beforetheB-bandmaximumand photons . The value of κ is expected to lie in the range
γ
quickly drops to less than 5% around and after the B- between 0.025 and 0.033cm2g−1 (Swartz et al. 1995).
band maximum. Given the inference that the SN might We adopt a value of 0.025cm2g−1 for the optically thin
be UV-luminous before maximum and the observational
regime. The form factor q describes the distribution of
fact that the SN is among the UV-bright Type Ia SNe 56Ni and thus 56Co (Jeffery 1999). For evenly mixed
aroundmaximum,ourcalculatedcorrectionprobablyun-
56Ni, the value of q is close to one-third. Taking element
derestimates the UV radiation. Using the Swift data
stratification and mixing in the interfaces into account,
around maximum, we estimate that this UV correction
Scalzo et al. (2014a) found that q = 0.45±0.05. Here,
introduces a systematic uncertainty of a few percent to
we adopt q =0.45 in our estimation.
the bolometric luminosity. Around maximum when the
Thevalueofv canbeobtainedbyconservationofen-
SN cools down, the UV contribution to the bolometric e
ergy. The total kinetic energy of the ejecta is 6M v2.
luminosity becomes even less important. ej e
Neglecting the radiation energy, the total kinetic energy
In the IR, the correction above 9700˚A is below 10%
is equal to the difference between the nuclear energy re-
around the B-band maximum, and then reaches a maxi-
leased in the explosion and the binding energy of the
mumof24%aroundthesecondarymaximuminthenear-
exploding WD. The binding energy of a rotating WD
IR. At the epochs with IR data, we find that the calcu-
with mass M and central density ρ is given in Yoon
lated correction is consistent with the IR measurements. ej c
& Langer (2005). Here we restrict the central density to
The final bolometric light curve is shown in Figure 13.
lie between 107 and 1010gcm−3.
WefurtheremployGaussianprocessregressiontoderive
Ifwefurtherassumethattheejectaiscomposedofun-
a maximum bolometric luminosity L =(2.2±0.2)×
max burned CO and synthesized Si, Fe, and Ni, then the nu-
1043ergs−1 at −2.6 days.
clear energy of the SN explosion is formulated in Maeda
& Iwamoto (2009) as a function of mass M and mass
4.2. 56Ni Mass and Ejecta Mass ej
fractions f , f and f . The ratio η =f /(f +f )
Fe Ni Si Ni Ni Fe
Next, we follow the procedure in Scalzo et al. (2012, is also a function of ρ . Following Scalzo et al. (2014a),
c
2014a)toderivethe56Nimassandthetotalejectamass. we adopt a Gaussian prior
First, the 56Ni mass can be estimated through the fol-
η =0.95−0.05ρ ±0.03max(1,ρ ) , (5)
lowing equation c,9 c,9
where ρ is ρ in units of 109gcm−3. In addition, we
L =αS(t ) , (1) c,9 c
max R restrict the mass fraction f less than 10%.
CO
where S(t ) is the instantaneous radioactive power at Based on the above assumptions, with a given set of
R
the bolometric luminosity maximum. α is an efficiency ejecta mass M , central density ρ and the mass frac-
ej c
10 Cao et al.
tions of different elements, we can calculate the maxi-
mum bolometric luminosity and t0 in equation (3) and 18 1e 167
compare them with our measurements of iPTF13asv. 19 uvm2
HereweperformMarkov-ChainMonte-Carlosimulations 20 6
for one million steps and obtain M = 0.81+0.10 and 21
Ni −0.18
Mej =1.44+−00..4142M(cid:12) at a 95% confidence level. ude 2232 51)−
gnit 24 42m
iSncIasnluzpooerdre-4teCr.3aht.lao.nD(ed2exr0tpaa1slc2aeh)iknehdhatyrhSpeheoevtalehllnmeStsosuiz,sretrSoccauaonlnszdtsoiatnategintottnahaSle.riy(SI2IN0shv1e0ell)olacdintedy- Apparent ma 22210519 uvw2 231Flux (ergsc−
tachedfromtheejecta. Theshellisacceleratedtoacon- 22
stantspeedv bycollidingwithfast-movingejectawith 23 -13.7 days 1
sh
24
velocities greater than v . In fact, some simulations
sh 25 0
of WD mergers show that the outermost material forms 20 10 0 10 20 3000 3500 4000 4500 5000
suchastationaryenvelopethatcollideswithfast-moving Days since B-band max Wavelength ( )
ejecta (Hoeflich & Khokhlov 1996). Following the calcu- Figure 14. Comparison between the SN-companion interaction
lation procedure in Scalzo et al. (2010) and Scalzo et al. signature model and iPTF13asv data. Left: the uvm2 and uvw2
lightcurvesofiPTF13asv(blackcircles)arecomparedtothetotal
(2012), we derive an envelope mass of 0.15+−00..1011M(cid:12) for lightcurves (green) which combine the SN-companion interaction
iPTF13asv. Thisshellincreasesthetotalmassofthesys- component(red;Kasen2010)andtheSNintrinsicemission(cyan).
temto1.59+0.45M The56Ni, shell, andtotalmassesof Right: theiPTF13asvspectrumat−13.7days(black)iscompared
−0.12 (cid:12) tothethermalspectrumofSN-companioncollision(red).
iPTF13asv are similar to those derived for SN20080522-
000 in Scalzo et al. (2012).
Thesetwoobservationalfacts, togetherwiththenear-IR
Thedetachedshellhaslittleeffectontheγ-rayopacity
secondary peak, strongly suggest that iPTF13asv has a
andpeakluminosityofanSN.Sincetherisetimeispro-
stratified ejecta along the line of sight, with strong con-
portional to M1/2, the massless detached shell will not centration of iron group elements near the center of the
tot
make the SN rise substantially longer than usual. explosion.
5. DISCUSSIONS 5.2. iPTF13asv as an Intermediate Case between
5.1. Origin of Strong UV Emission Normal and Super-Chandrasekhar Subclasses
StrongUVemissioninanSNIamaybepoweredbyan InTable4,wesummarizeacomparisonofnormalSNe,
extrinsic SN-companion collision (Kasen 2010). In fact, iPTF13asv, and super-Chandrasekhar SNe. As can be
in the UV-luminous SN2011de, Brown (2014) offered a seen from the table, on the one hand, iPTF13asv shares
possible explanation for its Swift light curve as a colli- similar light curve shapes and near-IR secondary peak
sion between the SN ejecta and a companion star. Here with normal events. SNID also finds decent spectral
we consider the same model to interpret the strong UV matches between iPTF13asv and normal events. On
emission of iPTF13asv. the other hand, the peak radiation of iPTF13asv is as
We utilize the scaling relation in Kasen (2010) to fit bright as super-Chandrasekhar events in both optical
the observed uvm2 light curve. In order to account for and UV. The evolution of Si II velocities of iPTF13asv
the non-negligible emission from the SN itself, we use is also similar to those of super-Chandrasekhar events.
thewell-sampleduvm2lightcurveofSN2011feasatem- In addition, we derived an total ejecta mass slightly be-
plate. The fitting result shows that the companion star yond the Chandrasekhar mass limit. Hence, we classify
is located at 2×1013cm away from the exploding WD iPTF13asv as an intermediate case between normal and
(left panel of Figure 14). Given a typical mass ratio of a super-Chandrasekhar subclasses.
few, the companion has a radius of ∼100R and fills its In addition to the features listed in the table, the H-
(cid:12)
Roche lobe. band break, a sharp spectral feature formed by absorp-
Although the model fit to the UV light curve looks tion of Fe II, Co II and Ni II (Hsiao et al. 2013), is
plausible, it overpredicts the SN emission at very early also distinctive between super-Chandrasekhar and nor-
phases. IntheR-bandlightcurvewithinafewdaysofex- malevents. TheH-bandbreakemergesaroundthemax-
plosion, the model-predicted SN flux (200µJy) is higher imum for normal SNe and decays to disappear within
than the observed fluxes (the inset of Figure 1) by a fac- a month of maximum. In contrast, this feature does
tor of > 30%. At −13.7 days, the predicted thermal not appear in the super-Chandrasekhar events. How-
emission flux from the model below 4000˚A is also much ever, the only near-IR spectrum of iPTF13asv is taken
one month after the maximum. Therefore, we cannot
higher than the observed spectrum (right panel of Fig-
determine whether iPTF13asv shows the H-band break
ure14). Hence,weconcludethatthestrongUVemission
or not.
seeniniPTF13asvisnotproducedbySN-companioncol-
lision.
5.3. Progenitor
Asaresultoftheaboveanalysis, weareforcedtocon-
clude that the strong UV emission is intrinsic. In fact, The massive ejecta and the stratification of the ejecta
the strong UV emission and the lack of iron in early- favor a DD progenitor system for iPTF13asv. In an SD
phase spectra are probably causally related, as the iron system, a non-rotating WD cannot exceed the Chan-
group elements are the major absorbers of UV photons. drasekhar mass limit, and it is not clear in reality how