Slowly fading super-luminous supernovae that are not pair-instability explosions

Article metrics

  • A Corrigendum to this article was published on 05 October 2016


Super-luminous supernovae1,2,3,4 that radiate more than 1044 ergs per second at their peak luminosity have recently been discovered in faint galaxies at redshifts of 0.1–4. Some evolve slowly, resembling models of ‘pair-instability’ supernovae5,6. Such models involve stars with original masses 140–260 times that of the Sun that now have carbon–oxygen cores of 65–130 solar masses. In these stars, the photons that prevent gravitational collapse are converted to electron–positron pairs, causing rapid contraction and thermonuclear explosions. Many solar masses of 56Ni are synthesized; this isotope decays to 56Fe via 56Co, powering bright light curves7,8. Such massive progenitors are expected to have formed from metal-poor gas in the early Universe9. Recently, supernova 2007bi in a galaxy at redshift 0.127 (about 12 billion years after the Big Bang) with a metallicity one-third that of the Sun was observed to look like a fading pair-instability supernova1,10. Here we report observations of two slow-to-fade super-luminous supernovae that show relatively fast rise times and blue colours, which are incompatible with pair-instability models. Their late-time light-curve and spectral similarities to supernova 2007bi call the nature of that event into question. Our early spectra closely resemble typical fast-declining super-luminous supernovae2,11,12, which are not powered by radioactivity. Modelling our observations with 10–16 solar masses of magnetar-energized13,14 ejecta demonstrates the possibility of a common explosion mechanism. The lack of unambiguous nearby pair-instability events suggests that their local rate of occurrence is less than 6 × 10−6 times that of the core-collapse rate.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Optical light curves of slow-fading super-luminous supernovae.
Figure 2: Spectral evolution of PTF 12dam and PS1-11ap from super-luminous supernovae of type I to SN 2007bi-like.
Figure 3: Spectral comparison with pair-instability and magnetar-driven supernova models.
Figure 4: Bolometric light curve and magnetar fit.


  1. 1

    Gal-Yam, A. et al. Supernova 2007bi as a pair-instability explosion. Nature 462, 624–627 (2009)

  2. 2

    Quimby, R. M. et al. Hydrogen-poor super-luminous stellar explosions. Nature 474, 487–489 (2011)

  3. 3

    Chomiuk, L. et al. Pan-STARRS1 discovery of two ultraluminous supernovae at z0.9. Astrophys. J. 743, 114–132 (2011)

  4. 4

    Cooke, J. et al. Super-luminous supernovae at redshifts of 2.05 and 3.90. Nature 491, 228–231 (2012)

  5. 5

    Barkat, Z., Rakavy, G. & Sack, N. Dynamics of supernova explosion resulting from pair formation. Phys. Rev. Lett. 18, 379–381 (1967)

  6. 6

    Rakavy, G. & Shaviv, G. Instabilities in highly evolved stellar models. Astrophys. J. 148, 803–816 (1967)

  7. 7

    Kasen, D., Woosley, S. E. & Heger, A. Pair instability supernovae: light curves, spectra, and shock breakout. Astrophys. J. 734, 102 (2011)

  8. 8

    Dessart, L., Waldman, R., Livne, E., Hillier, D. J. & Blondin, S. Radiative properties of pair-instability supernova explosions. Mon. Not. R. Astron. Soc. 428, 3227–3251 (2013)

  9. 9

    Heger, A. & Woosley, S. E. The nucleosynthetic signature of population III. Astrophys. J. 567, 532–543 (2002)

  10. 10

    Young, D. R. et al. Two type Ic supernovae in low-metallicity, dwarf galaxies: diversity of explosions. Astron. Astrophys. 512, A70 (2010)

  11. 11

    Pastorello, A. et al. Ultra-bright optical transients are linked with type Ic supernovae. Astrophys. J. 724, L16–L21 (2010)

  12. 12

    Gal-Yam, A. Luminous supernovae. Science 337, 927–932 (2012)

  13. 13

    Dessart, L., Hillier, D. J., Waldman, R., Livne, E. & Blondin, S. Super-luminous supernovae: 56Ni power versus magnetar radiation. Mon. Not. R. Astron. Soc. 426, L76–L80 (2012)

  14. 14

    Kasen, D. & Bildsten, L. Supernova light curves powered by young magnetars. Astrophys. J. 717, 245–249 (2010)

  15. 15

    Quimby, R. M. et al. Discovery of a super-luminous supernova, PTF12dam. Astron. Tel. 4121, (2012)

  16. 16

    Thompson, T. A., Chang, P. & Quataert, E. Magnetar spin-down, hyperenergetic supernovae, and gamma-ray bursts. Astrophys. J. 611, 380–393 (2004)

  17. 17

    Woosley, S. E. Bright supernovae from magnetar birth. Astrophys. J. 719, L204–L207 (2010)

  18. 18

    Umeda, H. & Nomoto, K. How much 56Ni can be produced in core-collapse supernovae? Evolution and explosion of 30–100 stars. Astrophys. J. 673, 1014–1022 (2008)

  19. 19

    Jerkstrand, A., Fransson, C. & Kozma, C. The 44Ti-powered spectrum of SN 1987A. Astron. Astrophys. 530, A45 (2011)

  20. 20

    Moriya, T. et al. A core-collapse model for the extremely luminous type Ic SN2007bi: an alternative to the pair-instability supernova model. Astrophys. J. 717, L83–L86 (2010)

  21. 21

    Inserra, C. et al. Super-luminous Ic supernovae: catching a magnetar by the tail. Astrophys. J. 770, 128 (2013)

  22. 22

    Arnett, W. D. Type I supernovae. I — Analytic solutions for the early part of the light curve. Astrophys. J. 253, 785–797 (1982)

  23. 23

    Tonry, J. L. et al. First results from Pan-STARRS1: faint, high proper motion white dwarfs in the Medium-Deep fields. Astrophys. J. 745, 42 (2011)

  24. 24

    Berger, E. et al. Ultraluminous supernovae as a new probe of the interstellar medium in distant galaxies. Astrophys. J. 755, L29 (2012)

  25. 25

    Young, D. R. et al. Core-collapse supernovae in low-metallicity environments and future all-sky transient surveys. Astron. Astrophys. 489, 359–375 (2008)

  26. 26

    Dahlen, T. et al. High-redshift supernova rates. Astrophys. J. 613, 189–199 (2004)

  27. 27

    Kobayashi, C., Tominaga, N. & Nomoto, K. Chemical enrichment in the carbon enhanced damped Lyα system by population III supernovae. Astrophys. J. 730, L14 (2011)

  28. 28

    Drout, M. R. et al. The first systematic study of type Ibc supernova multi-band light curves. Astrophys. J. 741, 97 (2011)

  29. 29

    Valenti, S. et al. A spectroscopically normal type Ic supernova from a very massive progenitor. Astrophys. J. 749, L28 (2012)

  30. 30

    Mazzali, P. A. et al. Models for the type Ic hypernova SN 2003lw associated with GRB 031203. Astrophys. J. 645, 1323 (2006)

  31. 31

    McCrum, M. et al. The super-luminous supernova PS1-11ap: bridging the gap between low and high redshift. Mon. Not. R. Astron. Soc (submitted)

Download references


We thank D. Kasen and L. Dessart for sending us their model data. The Pan-STARRS1 Surveys (PS1) have been made possible through contributions of the Institute for Astronomy, the University of Hawaii, the Pan-STARRS Project Office, the Max Planck Society (and its participating institutes, the Max Planck Institute for Astronomy, Heidelberg, and the Max Planck Institute for Extraterrestrial Physics, Garching), The Johns Hopkins University, Durham University, the University of Edinburgh, Queen’s University Belfast, the Harvard-Smithsonian Center for Astrophysics, the Las Cumbres Observatory Global Telescope Network Incorporated, the National Central University of Taiwan, the Space Telescope Science Institute, NASA grant no. NNX08AR22G issued through the Planetary Science Division of the NASA Science Mission Directorate, National Science Foundation grant no. AST-1238877, and the University of Maryland. S.J.S. acknowledges FP7/2007-2013/ERC Grant agreement no. 291222; J.L.T. and R. P. Kirshner acknowledge NSF grants AST-1009749, AST-121196; G.L. acknowledges Swedish Research Council grant no. 623-2011-7117; A.P., L.T., E.C., S.B. and M.T.B. acknowledge PRIN-INAF 2011. Work is based on observations made with the following telescopes: the William Herschel Telescope, Gran Telescopio Canarias, the Nordic Optical Telescope, Telescopio Nazionale Galileo, the Liverpool Telescope, the Gemini Observatory, the Faulkes North Telescope, the Asiago Copernico Telescope and the United Kingdom Infrared Telescope.

Author information

M.N. carried out the optical and near-infrared photometric and spectroscopic data analysis and wrote the manuscript. S.J.S. initiated, coordinated and managed the project, and contributed to manuscript preparation. A.J. carried out the theoretical modelling aspects, with contributions from S.A.S. C.I. reduced the ultraviolet data and assisted in all aspects of the analysis, including writing software to determine k-corrections and bolometric luminosity and running line identification routines. M.McC. provided the PS1-11ap reduced data. M.F. and R.K. carried out observations and coordinated Liverpool Telescope and Faulkes Telescope data. D.W., T.-W.C., K.S., D.R.Y., S.V., M.T.B., M.F., R.K. and Y.U. worked on finding PS1 transients using manual searching and software development. D.R.Y. wrote and adapted the Monte Carlo code described. F.B. and R. P. Kudritzky provided Gemini data through joint programmes. D.A.H. provided data from Faulkes North Telescope. A.P., L.T., E.C. and S.B. undertook observations with the Asiago telescopes. S.M., E.K., T.K., G.L., J.S. and F.T. provided data and relevant reductions through their Nordic Optical Telescope programmes. E.B., R.C., G.N., R.J.F., A.R., S.R., A.G.R., D.S., S.G., S.R., W.M.W.-V., N.S., R.M., R.L., A.S., D.M. and R. P. Kirshner worked on PS1 data analysis including difference imaging for PS1-11ap through the photpipe software at CfA/JHU and ensuring difference images were photometrically calibrated, and manual searching and spectroscopic follow-up of PS1 transients. N.E.-R., A.M.-G. and S.T. provided and reduced the GTC spectral data. J.L.T., M.E.H., W.S.B., K.C., H.A.F., E.A.M., N.K., N.M., J.M., P.A.P., C.W.S., W.S. and C.W. worked on designing and operating the PS1 system, from hardware through to software and data reduction routines.

Correspondence to M. Nicholl.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 Multi-colour photometry of PTF 12dam.

Observed light curve of PTF 12dam in UVW2, UVM2, UVW1, u, g, r, i, z (AB magnitudes) and J, H, K (Vega magnitude system).

Extended Data Figure 2 Image subtraction for the three earliest Pan-STARRS1 epochs of PTF 12dam in gP1, rP1 and iP1, using SDSS frames as reference images (taken on 11 February 2003).

These illustrate reliable image subtraction, resulting in clear detections of PTF 12dam at early phases. The images on the left are our PS1 detections, those in the centre are the SDSS templates, and on the right are the differences between the two. The bright star in the lower right was saturated and hence does not subtract cleanly. At each PS1 epoch there are two images, taken as transient time interval pairs. Photometry was carried out and determined in the SDSS photometric system to match the bulk of the follow-up griz imaging. The white areas are gaps between the 590 × 598 pixel cells in the PS1 chip arrays.

Extended Data Figure 3 Spectral evolution of PTF 12dam.

Full time-series optical and near-infrared spectroscopy of PTF 12dam, from two weeks before maximum light to an extended pseudo-nebular phase at 100 to >200 days afterwards. A Starburst99 model continuum spectral energy distribution for the host galaxy has been calibrated against SDSS and GALEX (Galaxy Evolution Explorer) photometry and subtracted from the last three spectra. RF, rest-frame.

Extended Data Figure 4 Effective temperature evolution of PTF 12dam and SN 2007bi, compared with magnetar-powered and pair-instability models.

The magnetar model comes much closer to reproducing the high photospheric temperatures we observe, and matches the gradient of the decline phase well. PISN models do not reach such high effective temperatures, and show an approximately 100-day temperature plateau as they rise, before declining after maximum light.

Extended Data Figure 5 Modelling of the O i, Mg i and Fe ii line fluxes in SN 2007bi at 367 days post-peak.

We plot contours for oxygen, magnesium and iron line fluxes predicted by our model in units of L = 1040 erg s−1 (dark blue = L/3; light blue = L; red = 3L; where L is the approximate luminosity of the lines in the 367-day post-peak spectrum of SN 2007bi) as functions of the respective ion density, {nO i, nMg i, nFe ii}, and electron density, ne, at 5,000 K (approximately the temperature derived for the iron zone from the relative strengths of iron lines). The panels for O i and Mg i show two lines (O i 6,300, 7,774 Å; Mg i 4,571, 5,180 Å), whereas Fe ii shows only contours for the 5,200 Å blend. No blending is likely to occur for any of the oxygen lines; the region where they intersect therefore gives the allowed densities, constraining ne to about 107 cm−3 (this is quite insensitive to the temperature we assume). Blending is also unlikely for Mg i] 4,571 Å, and the allowed Mg i density is therefore the intersection of this contour with ne ≈ 107 cm−3, which can be seen to give nMg i 103 cm−3. At this magnesium density, we see that the Mg i 5,180 Å line makes some contribution to the 5,200 Å flux. Also shown is the allowed Fe ii density at this temperature, for iron-zone electron densities spanning a factor of ten either side of that in the oxygen/magnesium zones.

Extended Data Figure 6 Fits to the observed bolometric light curve of PTF 12dam with radioactive 56Ni powered ejecta.

The formal fits of the models with kinetic energies of 1052 and 1053 erg are good (see graph), but the required combinations of 56Ni masses and ejecta masses (see data table) are not produced in physical models; such large nickel fractions are only expected to be produced in thermonuclear explosions (supernova Ia or possibly PISN), whereas the total ejected mass corresponds to the core-collapse of a massive star below the pair-instability threshold.

Extended Data Table 1 Optical photometry of PTF 12dam in SDSS griz bands, and k-corrections derived from our spectra.
Extended Data Table 2 Photometry of PTF 12dam outside the optical range.
Extended Data Table 3 Pan-STARRS1 photometry of PS1-11ap used in this work.
Extended Data Table 4 Log of spectra for PTF 12dam and the PS1-11ap spectra used in this work.

Supplementary information

Supplementary Information

This file contains Supplementary Sections 1-6. Sections 1-3 show data acquisition, reduction and analysis, Section 4 contains spectroscopic modelling, Section 5 light curve modelling and Section 6 local PISN rate calculation and further discussion on PISN searches. (PDF 864 kb)

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Nicholl, M., Smartt, S., Jerkstrand, A. et al. Slowly fading super-luminous supernovae that are not pair-instability explosions. Nature 502, 346–349 (2013) doi:10.1038/nature12569

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.