B. Dilday
1,2∗
, D. A. Howell
1,2
, S. B. Cenko
3
, J. M. Silverman
3
, P. E. Nugent
3,4
,
M. Sullivan
5
, S. Ben-Ami
6
, L. Bildsten
2,7
, M. Bolte
8
, M. Endl
9
, A. V. Filippenko
3
,
O. Gnat
10
, A. Horesh
12
, E. Hsiao
4,11
, M. M. Kasliwal
12,13
, D. Kirkman
14
,
K. Maguire
5
, G. W. Marcy
3
, K. Moore
2
, Y. Pan
5
, J. T. Parrent
1,15
,
P. Podsiadlowski
5
, R. M. Quimby
16
, A. Sternberg
17
, N. Suzuki
4
, D. R. Tytler
14
,
D. Xu
6
, J. S. Bloom
3
, A. Gal-Yam
18
, I. M. Hook
5
,
S. R. Kulkarni
12
, N. M. Law
19
, E. O. Ofek
18
, D. Polishook
20
, D. Poznanski
21
1
Las Cumbres Observatory Global Telescope Network, 6740 Cortona Dr., Suite 102, Goleta, California 93117, USA
2
Department of Physics, University of California, Santa Barbara, Broida Hall, Mail Code 9530, Santa Barbara, California 93106-9530, USA
3
Department of Astronomy, University of California, Berkeley, CA 94720-3411, USA
4
Lawrence Berkeley National Laboratory, Mail Stop 50B-4206, 1 Cyclotron Road, Berkeley, California 94720, USA
5
Department of Physics (Astrophysics), University of Oxford, Keble Road, Oxford, OX1 3RH, UK
6
Department of Particle Physics and Astrophysics, The Weizmann Institute of Science, Rehovot 76100, Israel
7
Kavli Institute for Theoretical Physics, University of California, Santa Barbara, CA, 93106, USA
8
UCO/Lick Observatory, University of California, Santa Cruz, California 95064, USA
9
McDonald Observatory, The University of Texas at Austin, Austin, TX 78712, USA
10
Racah Institute of Physics, The Hebrew University of Jerusalem, 91904, Israel
11
Carnegie Institution of Washington, Las Campanas Observatory, Colina El Pino, Casilla 601, Chile
12
Cahill Center for Astrophysics, California Institute of Technology, Pasadena, CA, 91125, USA
13
Observatories of the Carnegie Institution for Science, 813 Santa Barbara St, Pasadena CA 91101 USA
14
Center for Astrophysics and Space Sciences, University of California San Diego, La Jolla, CA 92093-0424, USA
15
Department of Physics and Astronomy, Dartmouth College, Hanover, NH, USA
16
IPMU, University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa-shi, Chiba, 277-8583, Japan
17
Minerva Fellow, Max Planck Institute for Astrophysics, Karl Schwarzschild St. 1, D-85741 Garching, Germany
18
Benoziyo Center for Astrophysics, The Weizmann Institute of Science, Rehovot 76100, Israel
19
University of Toronto, 50 St. George Street, Toronto M5S 3H4, Ontario, Canada
20
Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
21
School of Physics and Astronomy, Tel-Aviv University, Tel-Aviv 69978, Israel
∗
To whom correspondence should be addressed. E-mail: bdilday@lcogt.net
2
There is a consensus that Type-Ia supernovae (SNe Ia) arise from the ther-
monuclear explosion of white dwarf stars that accrete matter from a binary
companion. However, direct observation of SN Ia progenitors is lacking, and
the precise nature of the binary companion remains uncertain. A temporal se-
ries of high-resolution optical spectra of the SN Ia PTF 11kx reveals a complex
circumstellar environment that provides an unprecedentedly detailed view of
the progenitor system. Multiple shells of circumsteller are detected and the
SN ejecta are seen to interact with circumstellar material (CSM) starting 59
days after the explosion. These features are best described by a symbiotic nova
progenitor, similar to RS Ophiuchi.
Supernova PTF 11kx was discovered January 16, 2011 (UT) by the Palomar Transient Fac-
tory (PTF), at a cosmological redshift of z = 0.04660 ± 0.00001 (1). This corresponds to
a luminosity distance of ∼ 207 Mpc. The initial spectrum of PTF 11kx, taken January 26,
2011, showed saturated absorption in the Ca II H&K lines, along with weak absorption in the
Na I D lines (Fig. 1). In interstellar gas, the Na I D lines are usually of strength comparable to
(or greater than) that of the Ca II lines (2); thus, this combination of absorption features indi-
cated that PTF 11kx may have affected its surroundings, revealing evidence of its circumstellar
environment and progenitor system. SNe Ia are known to exhibit photometric (3, 4) and spec-
troscopic (5) diversity that is correlated with intrinsic brightness, such that bright (faint) SNe Ia
have relatively broad (narrow) light curves, high (low) photospheric temperature, and weak
(strong) Si II λ6150 absorption. Aside from the saturated Ca II absorption, the initial spectrum
and a subsequent temporal series of spectra show PTF 11kx to resemble SN 1999aa (6, 7), a
broad/bright SN Ia (Fig. 1). This suggests that insights into the progenitor of PTF 11kx may
be applicable to SNe Ia generally, rather than to only a subset of peculiar objects. A complete
sample of nearby SNe Ia shows that the subclass similar to SN 1999aa comprises ≥ 9% of all
3
SNe Ia (8).
We obtained high-resolution spectra (R ≈ 48, 000) of PTF 11kx with the Keck I High
Resolution Echelle Spectrometer (HIRES) instrument at −1, +9, +20, and +44 days, relative
to B-band maximum light. Low-resolution spectra were obtained on 14 epochs between −3
and +130 days and show the composition and evolution of features that are unique to SNe Ia
(Fig. 1). The SN system is blueshifted from the galaxy cosmological redshift by ∼ 100 km
s
−1
, as determined from the strongest narrow interstellar Na D absorption line (9, 10). An
assumption that emission lines from a nearby H II region evident in the two-dimensional spectra
are indicative of the progenitor velocity gives a consistent result.
The high-resolution spectra reveal narrow absorption lines of Na I, Fe II, Ti II, and He I
(Fig. 2). All are blueshifted by a velocity of ∼ 65 km s
−1
relative to the progenitor system,
with a velocity dispersion of ∼ 10 km s
−1
, and are consistent with arising in the same cloud or
shell. The sodium lines increase in depth over time, a feature which has been resolved in high-
resolution spectra of two previous SNe Ia (SN 2006X and SN 2007le), and has been explained as
the photoionization of CSM by the SN light, followed by recombination that is detected through
the increased presence of neutral sodium (9,11). A statistical study using high-resolution spectra
of 35 SNe Ia has shown that at least 20% of those with spiral host galaxies have circumstellar
material revealed through narrow Na I D absorption lines (10).
However, narrow lines of circumstellar Fe II, Ti II, and He I have not been seen previously
in a SN Ia. He I λ5876 strengthens over time, which can be explained by photoionization and
recombination, if sufficient far-ultraviolet (UV) photons are generated in the early phases of the
SN. SNe Ia are expected to produce only weak emission in the far-UV, but interaction of the SN
ejecta with an extended progenitor such as a red giant star can produce an excess of X-ray and
UV photons (12). Lower excitation states of Fe II λλ4923, 5018, 5169 decrease in strength in
time, while higher excitation states such as Fe II λ5316 remain constant. A possible explanation
4
for this is that the excitation energy of Fe II λ5316 (3.153 eV) corresponds to the saturated
Ca II K line, and thus the population of this level is unaffected by the SN light. Because SNe Ia
are weak sources in the UV, the distance over which they can ionize gas is limited, and hence
the observed photoionization indicates that the material is circumstellar (9, 11). An alternative
explanation, as the projection effect of different interstellar clouds, was proposed for the time-
variable Na I D lines in SN 2006X (13); however, interaction with the CSM in PTF 11kx
demonstrates conclusively that CSM is present. There are other narrow sodium features at
different relative velocities that do not vary with time, which must originate at a larger distance
and are most likely interstellar (Fig. 2).
The hydrogen Balmer series is clearly detected in absorption at ∼ 65 km s
−1
, and is consis-
tent with being at the same velocity as the Fe II, Ti II, Na I, and He I lines (Fig. 2). The Hα
and Hβ lines show narrow P-Cygni profiles, which are characteristic of an expanding shell of
radiating material. The presence of circumstellar hydrogen has long been identified as one of
the characteristics expected in the single-degenerate SN Ia progenitor model (14, 15), but has
only been detected in at most two other cases, SN 2002ic (16) and SN 2005gj (17), and only at
levels much stronger than in the early epochs of PTF 11kx.
These narrow features of H, He, Fe, Ti, and Na arise from material that is distinct from the
stronger Ca absorption, which, owing to its more blueshifted absorption minimum of ∼ 100 km
s
−1
, must be at a higher velocity. In fact, at later epochs, emission lines of Ca and H emerge,
marking the onset of interaction between SN ejecta and CSM. This transition of circumstellar
lines from absorption to emission has not previously been observed in a SN Ia, and shows
definitively the presence of CSM. Narrow absorption components can be seen atop the broader
emission (FWHM ≈ 1000 km s
−1
; Fig. 3), indicating that there are two regimes of material,
with the faster-moving H and Ca interior to the slower-moving shell (Fig. 4.)
The saturation of the Ca II K line near maximum light requires that no part of the SN
5
