Ultra-compact X-ray binaries (UCXBs) are low-mass X-ray binaries with ultra-short orbital periods (usually less than 1 hr) and hydrogen-poor donor stars, which are proposed to be potential Laser Interferometer Space Antenna (LISA) sources. In this work, we first employ the Modules for Experiments in Stellar Astrophysics code to examine the parameter space of the progenitors of UCXBs that LISA will detect. Our simulations indicate that the initial binaries with a neutron star and a companion star, in an orbit of initial orbital period smaller than the bifurcation period, could evolve into UCXBs, some of which will emit gravitational wave signals that can be detectable by LISA. However, the initial orbital periods of the binaries that will evolve into UCXB-LISA sources in a distance of 10 kpc are located in a very narrow range, i.e., the formation of these LISA source requires an extreme fine-tuning of initial parameter. According to the characteristic strains and the derived maximum detectable distances, four sources among eight UCXBs with the observed distances are expected to be detected by LISA. Based on the parameter space given by the detailed binary evolution models and the rapid binary star evolution code, the birthrate of UCXBs appearing as LISA sources in the Galaxy is estimated to be . Considering the contribution of UCXBs in globular clusters, the number of UCXB-LISA sources can reach 240–320. Although the formation condition is severe, the detectability of UCXBs by LISA is still significant because it provides an opportunity to pursue full multi-messenger investigations.

Ultracompact X-ray binaries (UCXBs) are low-mass X-ray binaries with hydrogen-deficient mass donors and ultrashort orbital periods. They have been suggested to be the potential Laser Interferometer Space Antenna (LISA) sources in the low-frequency region. Several channels for the formation of UCXBs have been proposed so far. In this paper, we carried out a systematic study on the He star donor channel, in which a neutron star (NS) accretes matter from a He main-sequence (MS) star through Roche lobe overflow, where the mass transfer is driven by the gravitational wave radiation. First, we followed the long-term evolution of the NS+He MS binaries by employing the stellar evolution code Modules for Experiments in Stellar Astrophysics (mesa ), and thereby obtained the initial parameter spaces for the production of UCXBs. We then used these results to perform a detailed binary population synthesis approach to obtain the Galactic rates of UCXBs through this channel. We estimate the Galactic rates of UCXBs appearing as LISA sources to be |${\sim} 3.1\!-\!11.9\, \rm Myr^{-1}$| through this channel, and the number of such UCXB-LISA sources in the Galaxy can reach about 1–26 calibrated by observations. This work indicates that the He star donor channel may contribute significantly to the Galactic UCXB formation rate. We found that the evolutionary tracks of UCXBs through this channel can account for the location of the five transient sources with relatively long orbital periods quite well. We also found that such UCXBs can be identified by their locations in the mass-transfer rate versus the orbital period diagram.

It is generally believed that ultracompact X-ray binaries (UCXBs) evolved from binaries consisting of a neutron star accreting from a low-mass white dwarf (WD) or helium star where mass transfer is driven by gravitational radiation. However, the standard WD evolutionary channel cannot produce the relatively long-period (40–60 minutes) UCXBs with a high time-averaged mass-transfer rate. In this work, we explore an alternative evolutionary route toward UCXBs, where the companions evolve from intermediate-mass Ap/Bp stars with an anomalously strong magnetic field (100–10,000 G). Including the magnetic braking caused by the coupling between the magnetic field and an irradiation-driven wind induced by the X-ray flux from the accreting component, we show that intermediate-mass X-ray binaries (IMXBs) can evolve into UCXBs. Using the MESA code, we have calculated evolutionary sequences for a large number of IMXBs. The simulated results indicate that, for a small wind-driving efficiency f = 10⁻⁵, the anomalous magnetic braking can drive IMXBs to an ultra-short period of 11 minutes. Comparing our simulated results with the observed parameters of 15 identified UCXBs, the anomalous magnetic braking evolutionary channel can account for the formation of seven and eight sources with f = 10⁻³, and 10⁻⁵, respectively. In particular, a relatively large value of f can fit three of the long-period, persistent sources with a high mass-transfer rate. Though the proportion of Ap/Bp stars in intermediate-mass stars is only 5%, the lifetime of the UCXB phase is ≳2 Gyr, producing a relatively high number of observable systems, making this an alternative evolutionary channel for the formation of UCXBs.

Recent discoveries of double neutron star (DNS) mergers and ultra-stripped supernovae (SNe) raise the questions of their origin and connection. We present the first 1D model of a DNS progenitor system that is calculated self-consistently until an ultra-stripped iron core collapse. We apply the MESA code starting from a post-common-envelope binary consisting of a 1.35 M _⊙ NS and a 3.20 M _⊙ zero-age main-sequence helium star and continue the modeling via Case BB Roche-lobe overflow until the infall velocity of the collapsing iron core exceeds 1000 km s⁻¹. The exploding star has a total mass of ∼1.90 M _⊙, consisting of a ∼0.29 M _⊙ He-rich envelope embedding a CO core of ∼1.61 M _⊙ and an iron-rich core of ∼1.50 M _⊙. The resulting second-born NS has an estimated mass of ∼1.44 M _⊙, and we discuss the fate of the post-SN system, as well as the mild recycling of the first-born NS. Depending on the initial conditions, this family of systems is anticipated to reproduce the DNS mergers detected by the LIGO network.

Recent investigations on the delay time of type Ia supernovae have set useful constraints on the progenitors of type Ia supernovae. Here we have calculated the evolution of close binaries consisting of a white dwarf and a main-sequence or subgiant companion. We assume that, once Roche lobe overflow occurs a small fraction of the lost mass from the system forms a circumbinary disk, which extracts the orbital angular momentum from the system through tidal torques. Our calculations indicate that the existence of circumbinary disk can enhance the mass transfer rate and cause secular orbital shrinkage. The white dwarf can grow in mass efficiently to trigger type Ia supernovae even with relatively low-mass ($\la 2 M_{\odot}$) donor stars. Thus this scenario suggest a new possible evolutionary channel to those type Ia supernovae with long delay time $\sim 1-3$ Gyr.Comment: accepted for publication in ApJ Letter