DISCOVERY AND EARLY MULTI-WAVELENGTH MEASUREMENTS OF THE ENERGETIC TYPE IC SUPERNOVA PTF12GZK: A MASSIVE-STAR EXPLOSION IN A DWARF HOST GALAXY

Type Ic CCSNe are heterogeneous (for a review on CCSNe, see Filippenko 1997). Their luminosity, ejected mass, and kinetic energy span over an order of magnitude, from the subluminous SN 2004aw to the overluminous SN 1998bw (Mazzali et al. 2009; Drout et al. 2011). The light-curve shapes of different events are also quite diverse. A subclass of SNe Ic whose spectra are characterized by broad lines (Type Ic-BL; prototype SN 1998bw) is the only one for which clear evidence of an association with gamma-ray bursts (GRBs) exists (GRB-SNe; see Woosley & Bloom 2006 for a review). Superluminous SNe (SLSNe) of Type Ic are even more powerful (Gal-Yam 2012 and references therein), but these probably result from a different physical mechanism. While SNe Ic are common in the center of high-metallicity galaxies (Anderson et al. 2012), SLSNe-I and broad-lined GRB-SNe tend to be found in dwarf hosts (e.g., Stanek et al. 2006; Modjaz et al. 2008; Arcavi et al. 2010), giving untargeted sky surveys an advantage over targeted surveys in detecting these types of cosmic explosions.

The Palomar Transient Factory (PTF; Law et al. 2009; Rau et al. 2009) is a wide-field untargeted sky survey which explores the transient optical sky. It uses the PTF CFH12k camera mounted on the Palomar 48 inch telescope (P48). PTF's short observing cadence and real-time capability (e.g., Gal-Yam et al. 2011) enables the discovery and study of SNe at early stages of the explosion. In this Letter we report the discovery and study of PTF12gzk, a peculiar SN Ic in a dwarf star-forming galaxy located at redshift z = 0.0137 (distance 57.8 Mpc, distance modulus 33.8 mag, assuming H₀ = 71 km s⁻¹ Mpc⁻¹).

PTF12gzk was discovered on 2012 July 24.3 (UT dates are used herein) at α(J2000) = 22^h12^m4153 and δ(J2000) = +00°30'431, in the dwarf galaxy SDSS J221241.53+003042.7 (within the SDSS Stripe 82 footprint).³² The discovery magnitude was 20.66 in the r band,³³ and it was not detected down to mag 21.6 (3σ) in previous PTF images obtained on July 19 (Ben-Ami et al. 2012).

Shortly after discovery, we initiated an extensive follow-up campaign, including our Hubble Space Telescope (HST) Target-of-Opportunity (ToO) program for STIS ultraviolet (UV) spectroscopy of a stripped-envelope SN (Cycle 19, GO-12530; PI Filippenko) and Swift X-ray and UV photometry (Cycle 8, PID 8110099; PI Kasliwal), the results of which are presented herein (see Figures 1 and 2 for photometry and spectroscopy, respectively). We also triggered radio and millimeter observations using the JVLA (program 12A-363; PI Horesh) and the Combined Array for Research in Millimeter-wave Astronomy (program 12A-c0945; PI Horesh); see A. Horesh et al. (in preparation).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

3.1. Photometry

3.2. Spectroscopy

Extensive spectroscopy of PTF12gzk was performed, and detailed analysis will be presented in a future paper. A selection of optical spectra is shown in Figure 2.³⁵

The first spectrum was obtained on July 27, revealing the object to be a peculiar SN I with a blue continuum, reminiscent of the blue continuum observed in the early-time spectra of SN 2006aj (Mazzali et al. 2006b). Initially it was difficult to classify the spectrum securely, but later optical spectra resemble those of SNe Ic, with an absence of prominent He i lines (Ben-Ami et al. 2012).

Classification with the help of SNID (Blondin & Tonry 2007) of the August 9 Lick 3 m/Kast spectrum suggests that the best fit is to the peculiar SN Ic SN 2004aw near peak brightness (Taubenberger et al. 2006). A good match is also obtained for PTF10qts, a Type Ic-BL SN, at ∼30 days after explosion. All spectra show prominent, broad absorption lines of Ca ii, Si ii, Fe ii, and O i, with SYNOW fits yielding maximum velocities of ∼35, 000 km s⁻¹ for the July 27 spectrum, and ∼20, 000 km s⁻¹ for the August 9 spectrum. See Figure 3 and Section 5 for further discussion.

Zoom In Zoom Out Reset image size

The absence of the 2.1 μm He i feature in the IR TSPEC spectrum taken on August 7 (Figure 3, bottom-right panel) suggests that He is not abundant in the envelope (Hachinger et al. 2012). While a strong absorption line is present at 1 μm, which could be identified with He i 1.0830 μm, in this case it is likely contaminated by other elements such as C, Mg, S, and Ca (e.g., Mazzali& Lucy 1998; Sauer et al. 2006; Hachinger et al. 2012).

The first HST UV spectrum, obtained by the Space Telescope Imaging Spectrometer with the near-UV Multi-Anode Microchannel Array (STIS/MAMA), was taken between 20:06 August 6 and 03:35 August 7, with a total exposure time of 11,278 s. We also obtained optical spectra with the STIS/CCD, using grisms centered at 4300 and 7500 Å. A mosaic of all three HST spectra is given in Figure 3, with a focus on the UV spectrum in the bottom-right panel. The flux deficit with respect to blackbody spectra seen in the STIS/MAMA UV spectrum is characteristic of all SNe I (Panagia 2007) and indicates strong line blanketing, evidence for a highly mixed outer envelope devoid of hydrogen. Full analysis of the HST UV spectra will be presented by S. Ben-Ami et al. (in preparation).

An image and a spectrum of the host galaxy obtained by the SDSS prior to explosion are shown in Figure 2. The SN exploded very close to the center of the host galaxy (offset 00 N and 04 E). We downloaded the host-galaxy spectrum from the SDSS 9th Data Release (DR9; Ahn et al. 2012). After correcting for the host-galaxy redshift, we measure integrated emission-line fluxes using standard procedures via splot in IRAF, and follow Pérez-Montero & Diaz (2003) to compute statistical errors.³⁶ Using the SDSS Petrosian magnitudes, correcting for Galactic and host-galaxy extinction (see below), and applying K-corrections via kcorrect (v4.2, Blanton & Roweis 2007) at the host-galaxy redshift, we derive an absolute magnitude of M_r = −14.8 mag. This is an extremely underluminous dwarf galaxy, well below the luminosities of the Large and Small Magellanic Clouds. It is one of the least luminous host galaxies of CCSNe discovered by PTF, and it is not a common host galaxy for an SN Ic, even considering those found via untargeted surveys (Arcavi et al. 2010), except for SNe Ic-BL. In addition, the clear detections of numerous bright emission lines allow us to apply standard diagnostics of the star formation rate (SFR), extinction, and metallicity to this galaxy.

4.1. Star Formation Properties

4.2. Oxygen Abundance

PTF12gzk is a luminous SN Ic, at the high end of the SN Ic luminosity distribution (Drout et al. 2011). It exhibits a slow rise of 18 days to its peak r-band magnitude, with B peaking ∼10 days earlier. This is a large gap relative to other SN I, though similar to SN 2004aw (Taubenberger et al. 2006); it is caused by metal-line absorption from heavy elements in the outer layers of the ejecta, as is evident from spectra taken after August 1.

A least-squares fit to a f(t)∝(t − t₀)ⁿ behavior of our well-sampled early photometry places the explosion date between 10 and 40 hr prior to our discovery at the 95% confidence level.³⁷ We cannot rule out the popular quadratic fireball model, but higher values of n are possible as well for larger areas in the fit parameter space (Figure 1, bottom panels).

Spectroscopically, PTF12gzk exhibits high expansion velocities, ∼30, 000 km s⁻¹ (Si ii absorption velocity). Other SNe Ic with similar velocities are broad-lined SNe Ic (Figure 4), some of which are associated with GRBs (Woosley & Bloom 2006 and references within), while no such association was determined for PTF12gzk (see also SN 2009bb; Soderberg et al. 2010). Most similar is SN 2003lw, an SN associated with a GRB (Mazzali et al. 2006a). A possible explanation is a burst misaligned with our line of sight, or a failed GRB. Such a scenario is further supported by the host-galaxy characteristics, resembling those of a broad-lined SN Ic host galaxy. We know of no typical SN Ic exploding in a host with similar luminosity and oxygen abundance. The observed relatively narrow lines give a dispersion of Δv/v ≈ 0.25, compared to ∼1 in the case of broad-lined SNe Ic, and may suggest a nonspherical explosion geometry (Leonard et al. 2006), or that the ejecta mass is high or has a very steep density gradient (a discussion on the effects of a steep density gradient on the light curve (LC) can be found in Piro & Nakar 2012). Late-time, nebular spectra will probe the geometry of the explosion in more detail.

Zoom In Zoom Out Reset image size

From the Si ii line velocity at peak brightness of PTF12gzk (15,300 km s⁻¹ from the August 12 spectrum) and SN 2004aw (12,400 km s⁻¹; J. Deng et al., in preparation), and the rise time of these two SNe, we use the following scaling relations (Arnett 1982; Mazzali et al. 2009; see also P. A. Mazzali et al., in preparation) to estimate the physical properties of PTF12gzk: τ ≈ κ^1/2M^3/4E^−1/4 and v = (2E/M)^1/2, where τ is the light-curve rise time, E is the kinetic energy, and κ is the opacity. The derived ejecta mass is 7.5 M_☉ (6–12 M_☉), pointing to a large initial progenitor mass of 25–35 M_☉, though the latter values are highly uncertain (Mazzali et al. 2000). We derive a kinetic energy of 7.5 × 10⁵¹ erg [(5–10) × 10⁵¹  erg]. Using L_max and t_max we get an estimated ⁵⁶Ni mass of 0.37 M_☉ (scaling the PTF12gzk light curve to that of SN 2003dh; Mazzali et al. 2003). Using the V-band peak magnitude versus nickel mass relation presented by Perets et al. (2010), we derive a ⁵⁶Ni mass of 0.35 M_☉, in agreement with the results derived from the scaling relations. These physical properties, as well as the high expansion velocities and the host galaxy, are unlike those of normal SNe Ic, which typically occur in large hosts and have low ejecta masses, and kinetic energies, (1.7 M_☉, and 10⁵¹ erg Drout et al. 2011. For nickel mass in Type Ic SNe see for example Taubenberger et al. 2006, and Sauer et al. 2006). Instead, they are reminiscent of GRB-SNe (Mazzali et al. 2009).

PTF12gzk is a remarkable example of an SN Ic in terms of expansion velocities, evolution timescale, the ejected mass, and the kinetic energy released in the explosion. We conclude that these properties point to the explosion of a massive star deficient in H and He, at the higher-mass end of SN Ic progenitors. This further illustrates the peculiar population of SNe Ic exploding in dwarf hosts (Arcavi et al. 2010), as seen also in the case of GRB-SNe and most SLSNe-I.

PTF12gzk demonstrates the advantages of using an untargeted sky survey such as PTF with an extensive network of instruments and telescopes in various wavebands to detect and rapidly characterize unusual cases of cosmic explosions.

S.B. is supported by a Ramon Fellowship from ISA. A.G. acknowledges support by grants from the ISF, BSF, GIF, Minerva, and the EU FP7/ERC. A.V.F. and his group benefit from financial assistance from Gary & Cynthia Bengier, the Richard & Rhoda Goldman Fund, the Sylvia & Jim Katzman Foundation, the Christopher R. Redlich Fund, the TABASGO Foundation, NSF grants AST-0908886 and AST-1211916, and NASA/HST grant GO-12530 from STScI (which is operated by the AURA, Inc., under NASA contract NAS 05-26555). P.A.M. and E.P. acknowledge financial support from grants INAF PRIN 2011 and ASI/INAF I/088/06/0. M.I. and Y.J. were supported by the Creative Initiative program of the NRFK. M.M.K. acknowledges Hubble and Carnegie-Princeton Fellowships. D.C.L. is supported by NSF grant AST-1009571. E.O.O. acknowledges the Arye Dissentshik career development chair and a grant from the Israeli MOST.

PTF is a collaboration of Caltech, LCOGT, the Weizmann Institute, LBNL/NERSC, Oxford, Columbia, IPAC, and UC Berkeley. We are grateful for the assistance of the staff at the various observatories used to obtain data. This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by JPL, Caltech, under contract with NASA.