Photospheric Velocity Gradients and Ejecta Masses of Hydrogen-poor Superluminous Supernovae -- Proxies for Distinguishing between Fast and Slow Events

Abstract

We present a study of 28 Type I superluminous supernovae (SLSNe) in the context of the ejecta mass and photospheric velocity. We combine photometry and spectroscopy to infer ejecta masses via the formalism of radiation diffusion equations. We show an improved method to determine the photospheric velocity by combining spectrum modeling and cross correlation techniques. We find that Type I SLSNe can be divided into two groups by their pre-maximum spectra. Members of the first group have the W-shaped absorption trough in their pre-maximum spectrum, usually identified as due to O II. This feature is absent in the spectra of supernovae in the second group, whose spectra are similar to SN~2015bn. We confirm that the pre- or near-maximum photospheric velocities correlate with the velocity gradients: faster evolving SLSNe have larger photosheric velocities around maximum. We classify the studied SLSNe into the Fast or the Slow evolving group by their estimated photosheric velocities, and find that all those objects that resemble to SN~2015bn belong to the Slow evolving class, while SLSNe showing the W-like absorption are represented in both Fast and Slow evolving groups. We estimate the ejecta masses of all objects in our sample, and obtain values in the range of 2.9 ($\pm$0.8) - 208 ($\pm$61) $M_\odot$, with a mean of $43 (\pm 12)~ M_\odot$. We conclude that Slow evolving SLSNe tend to have higher ejecta masses compared to the Fast ones. Our ejecta mass calculations suggests that SLSNe are caused by energetic explosions of very massive stars, irrespectively of the powering mechanism of the light curve.