Two types of Type II supernovae – the iron-core-collapse and the electron-capture supernovae – are thought to produce the majority of neutron stars in the Universe.
The studies of pre-supernovae evolution of stars have shown that there are exist 2 major types of supernova explosions: stars with masses > 10 Msun evolve to develop iron cores, which collapse resulting neutron stars and black holes, while stars with masses in range 6Msun-10Msun form electron degenerate carbon-oxygen cores and explode as carbon deflagration supernovae. However the third triggering mechanist was suggested by Rakavy eit al (1967) where the core collapses due to electron captures in stars with masses 8-10Msun that form O+Ne+Mg cores (Nomoto, 1984).These are called electron-capture supernovae.
Christian Knigge, Malcolm J. Coe, & Philipp Podsiadlowski in their recent Letter to Nature report that a large, X-ray pulsars hosted in Be/Xray binaries are actually can be distinguished into two distinct subpopulations based on their spin periods, orbital periods and orbital eccentricities.
Be/X-ray binaries contain a low mass neutron star and fast-rotating 8Msun-18Msun O9e-B2e main-sequence star surrounded by circumstellar decretion disk. The letter “e” in Be stands for emission, because the optical spectra of Be stars display emission lines including strong infrared emission. These emission lines and infrared excess are originated from the gaseous, equatorially concentrated circumstellar decretion disk which surrounds the OB star. This disc also constitutes the main source of variability in BeX.
BeXs are exceptionally abundant in the Small Magellanic Cloud (SMC). Actually the number of BeXs in SMC is comparable to the number of BeXs in the Milky Way, given the mass ratio of these galaxies 1:100 (SMC to Milky Way).
In the Large Magellanic Cloud (LMC), the number of BeXs is roughly in line with its stellar mass content and similar to found in Milky Way.
Analysis of orbital period (Porb), spin periods (Pspin) and orbital eccentricities of BeXs in SMC, LMC and Milky Way done by Christian Knigge, Malcolm J. Coe, & Philipp Podsiadlowski had shown that the log(Pspin) distribution of BeXs contains two distinct subpopulations with characteristic spin periods of Pspin ~ 10s and Pspin ~ 200s, where short-Pspinand long-Pspin subpopulations contribute about 35 and 65% to the total number, respectively. See the figure below:
|a, Distribution for all systems. b, c, Distribution broken down by host galaxy: SMC (b); Milky Way (MW) + LMC (c). All of these distributions are bimodal (modes shown in red and blue), and the double-Gaussian decomposition (black) suggested by the KMM algorithm25 (an algorithm that detects bimodality in an observational data set) is shown in each panel. The number of systems contributing to each observed distribution and the associated P value provided by the algorithm are also shown. Applying the KMM test to the subset of spectroscopically confirmed systems (not shown) gives P = 8 × 10−3 (N = 64). d, Direct comparison of the decompositions for the independent SMC and Milky Way + LMC populations, showing them to be mutually consistent. Figure credit of Christian Knigge, Malcolm J. Coe, & Philipp Podsiadlowski, Letters to Nature, 2011, doi:10.1038/nature10529|
These two subpopulations were shown to be highly statistically significant in the whole sample from 3 galaxies and also remain significant even if the data set is divided by host galaxy.
One of possible explanations for this trend is that two subpopulations of Be/Xray binaries can associated with the two types of Type II supernovae providing different mechanisms for neutron star formation where electron-capture supernovae are responsible for low-mass neutron stars in X-ray binaries with short spin periods, short orbital periods and low eccentricities, while iron-core-collapse supernovae produce high-eccentricity binaries containing high-mass neutron stars, with longer spin periods.