< Back to All Articles

SN 2008es: Strong Interacting Hydrogen-rich Superluminous Supernova without Narrow H-alpha Features and Early Dust Formation

K. Bhirombhakdi (kbhirombhakdi[at]stsci.edu)

Superluminous supernovae (SLSNe; Gal-Yam 2012, 2019) are core-collapse supernovae (CCSNe) that are 10–100 times brighter at peak than canonical supernovae (SNe). With peaks at magnitudes ≲–21 dex and very blue continuum, SLSNe can probe the early universe and inform us about, e.g., cosmology (Scovacricchi et al. 2016), star/galaxy evolution (Schulze et al. 2018), and the Population Ⅲ stars (Smidt et al. 2014).

Power origins of SLSNe are still uncertain. Proposed origins included, for example, efficient conversion of the core-collapse energy through strong circumstellar interaction (CSI), energy from a central engine such as a spinning-down magnetar (i.e., a strongly magnetized neutron star), fallback accretion onto a blackhole, or massive amount (i.e., ≳1 solar mass) of Ni56 (Wang et al. 2019). Currently, evidence supports that SLSNe are CCSNe, and are dominantly powered by either CSI or a spinning-down magnetar. In this newsletter, we briefly look at a recent study by Bhirombhakdi et al. 2019 which attempted to constrain the power source of the superluminous SN 2008es.

SN 2008es peaked at ~1044 erg/s (Gezari et al. 2009; Miller et al. 2009). It was classified as a hydrogen-rich (Type Ⅱ) event by showing strong and broad (i.e., ~10,000 km/s) H-alpha emission without P Cygni or narrow (i.e., ~100 km/s) feature. Since most hydrogen-rich SLSNe show strong narrow H-alpha features, and are dubbed SLSNe-Ⅱn ('n' for narrow H-alpha features similar to canonical SNe‑Ⅱn), SN 2008es was a special case, and SLSNe‑Ⅱ without narrow features are rare. Besides SN 2008es, other SLSNe‑Ⅱ without narrow features included, for example, SN 2013hx and PS15br (Inserra et al. 2018).

SLSNe/SNe Type Ⅱn are typically more luminous than their counterparts and, because of the narrow H-alpha signatures, are well-known for their strong CSI as the power sources. By lacking the signature, the power origin of SN 2008es was uncertain and both CSI and spinning-down magnetar models fit well to its light curve up to ~100 days after explosion (Inserra et al. 2018). In Bhirombhakdi et al. 2019, SN 2008es was observed extending to ~600 days in optical and near infrared (NIR). While both CSI and spinning-down magnetar models still fit well to the light curve, other evidence supported that SN 2008es was powered by strong CSI.

SN 2008es
Figure 1: Spectral energy distribution of SN 2008es. Data include gVRI (green) and K' (black) bands at 255 days, and HK' (red) bands at 300 days from Bhirombhakdi et al. 2019. Green solid line = 5,000 K blackbody scaled to R band representing the cooling down SN. Dashed lines = ~1,500 K blackbodies fit to 300-day HK' bands (red), and scaled to 255-day K' band (black). The estimated temperatures are with respect to the SN frame. Over the cooling down SN, the plot shows I excess due to strong H‑alpha emission, and NIR excess due to thermal dust.

First, shown in Figure 1, the 288-day H-alpha emission was strong. We measured the 288-day H-alpha luminosity ~1e40 erg/s, and its equivalent width ~800 Å, which is stronger than the 89-day H-alpha emission ~160 Å. These numbers are comparable to some strong interacting Type Ⅱn SLSNe/SNe at similar epochs (e.g., SN 1998S, Mauerhan & Smith 2012; SN 2006gy, Smith et al. 2010), supporting strong CSI in SN 2008es as well.

Second, the 288-day H-alpha profile was blue shifted with its blue wing extending to ~10,000 km/s, while the earlier-time H-alpha profile was more symmetric with broad wings on both sides ~10,000 km/s. This spectral feature was similar to some strong interacting Type Ⅱn SNe that formed dust at early times (i.e., ≲500 days after explosion) in cool-dense shells (CDSs; Fox et al. 2009, 2011). CDS is a region produced by thermal instability due to strong CSI, and supports dust condensation at early times (Chevalier & Fransson 2017). If formed, CDS dust can obscure emission from the red side causing red-wing attenuated (or blue-shifted) profile with constant blue-wing width. This is the feature we observed in SN 2008es.

SN 2008es Emission

Figure 2: H-alpha emission of SN 2008es at early-time 89 days (black; Miller et al. 2009) and later-time 288 days (red; Bhirombhakdi et al. 2019) after explosion. Fluxes were baseline subtracted, but uncalibrated. Savitzky-Golay filter from scipy package was applied for smoothing the spectra. The plot shows red-wing attenuation of H-alpha emission at 288 days relative to at 89 days. In both spectra, the blue wing still extended to ~10,000 km/s. This feature was interpreted as a result of CDS dust condensation due to strong CSI.

Moreover, CDS dust produces NIR excess with characteristic temperature ~1000 K at location associated to the shock front created by the explosion. We also observed this feature as presented in Figure 2. The figure shows gVRIHK' observations during 255—300 days after explosion. The optical component at 255 days (green) was modelled by a 5,000 K blackbody scaled to R band, representing the cooling down SN. Strong I excess was due to the strong H‑alpha emission. A NIR component at 300 days (red) was fit by a ~1,500 K blackbody with radius ~1e16 cm, which was associated to the shock front. After scaling this NIR component to the observed 255-day K' band (black), the NIR excess was evident compared to the optical component.

Last, by following analyses from Chevalier & Irwin (2011) and Moriya & Tominaga (2012), SN 2008es parameters were consistent with the case of a strong interacting SN without the narrow H‑alpha features. Its parameters implied that by the time of shock breakout circumstellar materials of SN 2008es were shocked to high velocities, leaving no slow-moving material to produce the narrow features.

In conclusion, the study observed SN 2008es behavior at later times and put a better constraint on its power origin. Multiple evidences supported that SN 2008es was powered by strong CSI extending to at least a year after explosion. The lacking of narrow H‑alpha emission was consistent with its parameters. However, we could not rule out the possibility of having a spinning-down pulsar powering scenario. We note that more samples of SLSNe‑Ⅱ without narrow features are necessary for better understanding of objects in this class. Additionally, SN 2008es was the first of its kind—SLSNe‑Ⅱ without narrow features—that we can observe CDS dust condensation. Together with other observed dust in SLSNe‑Ⅰ (e.g., iPTF16eh; Lunnan et al. 2018) and SLSNe‑Ⅱn (e.g., SN 2006gy; Miller et al. 2010), this completed the picture that SLSNe could be dusty at relatively early times around their peaks. Understanding dust in SLSNe would be important, especially if SLSNe would probe the early universe.


Bhirombhakdi, K., et al. 2019, MNRAS, 488, 3783

Chandra, P. 2018, SSRv, 214, 27

Chevalier, R. A., & Fransson, C. 2017, "Thermal and Non-thermal Emission from Circumstellar Interaction," in Handbook of Supernovae, eds. A. Alsabti and P. Murdin, Springer International Publishing AG, p. 875

Chevalier, R. A. & Irwin, C. M., 2011, ApJ, 729, L6

Fox, O., et al. 2009, ApJ, 691, 650

Fox, O., et al. 2011, ApJ, 741, 7

Gal-Yam, A. 2012, Science, 337, 927

Gal-Yam, A. 2019, ARA&A, 57, 305

Gezari, S., et al. 2009, ApJ, 690, 1313

Inserra, C., et al. 2018, MNRAS, 475, 1046

Lunnan, R., et al., 2018, NatAs, 2, 887

Mauerhan, J. & Smith, N. 2012, MNRAS, 424, 2659

Miller, A. A., et al. 2009, ApJ, 690, 1303

Miller, A. A., et al. 2010, AJ, 139, 2218

Moriya, T. J. & Tominaga, N. 2012, ApJ, 747, 118

Schulze, S., et al. 2018, MNRAS, 473, 1258

Scovacricchi, D. et al. 2016, MNRAS, 456, 1700

Smidt, J., et al. 2014, ApJ, 797, 97

Smith, N., et al. 2010, ApJ, 709, 856

Wang, S., Wang, L., & Dai, Z., 2019, arXiv:1902.07943