Research

Stars are the most widely recognized astronomical objects, and represent the most fundamental building blocks of galaxies. We know today that the majority of stars are members of binary or even higher order multiplicity systems. Binaries are excellent astrophysical laboratories, and compared to single isolated stars, they induce new processes, offering the opportunity to confront our understanding of a broad range of physics. The interactions between the stellar component of a binary can alter significantly the evolution of the system, and potentially lead to the formation of exotic objects and phenomena, from double compact objects in close orbits, to peculiar supernova events and gamma-ray bursts.

Phenomena related to binary evolution span scales from kilometers (e.g. the internal stellar structure), to gigaparsecs (e.g. stellar feedback processes in galaxy evolution). In my research so far, I dealt with problems related to all different scales, using in each case the adequate tools and level of detail. I studied the evolutionary history of Galactic X-ray binaries using detailed evolutionary models of both the internal structure of the stellar components, and the binary orbit. These studies allowed us to track the evolution of the X-ray binary back to the formation of the compact object, and study the properties of the binary during the supernova event. At the galactic scale, I worked on population synthesis studies of X-ray binaries at different galactic environments. The use of more approximate simulation tools, allowed me to perform a statistical comparison of my simulations’ results with observed extragalactic populations. Finally, at the largest scales, I combined binary population synthesis models with cosmological simulations, in order to study the evolution of X-ray binaries on cosmological timescales, and the effect that these populations have in galaxy formation and evolution, especially at high redshifts.

Bellow you can find more information about my current and former students and the different research projects I am working on.

Interests

  • Stellar and binary evolution
  • Formation and evolution of compact objects: black holes and neutron stars
  • Evolutionary history of Galactic X-ray binaries
  • Extragalactic populations of X-ray binaries
  • Evolution of triple and multiple stellar systems
  • Binary evolution in dense stellar systems
  • High performance computing in computational astrophysics

Students


Ying Qin

Ying Qin

Ph.D. student (01/2015 - Present). Ying's thesis is on the progenitors of long Gamma-ray bursts and high-mass X-ray binaries. Ying is supported through a national fellowship from the Chinese Scholarship Council.

Mads Sørensen

Mads Sørensen

Ph.D. student (09/2014 - Present). Mads' thesis is on the effects of binary stars on the rates of different classes of giant and Wolf-Rayet stars, the rate of different types of supernova, and the ionizing flux produced by a stellar population.

Michael Tremmel

Michael Tremmel

Former undergraduate student at Northwestern University. Michael's senior thesis was on on extragalactic X-ray sources and the evolution of X-ray binary populations across cosmic time. Using data from semi-analytical galaxy models, he simulated the X-ray binary populations of galaxies out to redshifts of around 10. Using these models, he calculated the integrated X-ray luminosities of these galaxies and derive galaxy X-ray luminosity functions to compare with observations done out to redshift of about 1.4. Michael is currently a Ph.D. student at the University of Washington.

Follow
Tsing-Wai Wong

Tsing-Wai Wong

Former pre-doctoral fellow at the Harvard-Smithsonian Center for Astrophysics (08/2012-08/2013): TsingWai's project was on modelling the evolutionary history of the high-mass X-ray binary IC10 X-1, which hosts a black hole. Tsing-Wai completed his Ph.D. on 12/2013 at Northwestern University.

Follow

Research Projects

  • The origin of stellar-mass black-hole spin

    The origin of stellar-mass black-hole spin

    The recent exciting developments in measuring the spins of stellar-mass black holes opened a new window for the study of compact objects and their formation mechanisms.In this research project, we study all the physical processes taking place in the evolution of a binary star, progenitor of X-ray binaries, that can potentially affect the spin of the resulting black hole.

    The recent exciting developments in measuring the spins of stellar-mass black holes opened a new window for the study of compact objects and their formation mechanisms. The evolution of isolated single massive stars with solar-like chemical abundances is not expected to lead to highly spinning black holes, like some of the observed ones. However, black holes are only observed in binary systems, and these new measurement suggest that the origin of the observed BH spins may closely connected to physical processes tied to binaries. At the same time, the measurement of stellar-mass black hole spins can be used as strong constraint to poorly understood physical processes taking place in the evolution of binary stars, such as the angular momentum transport between the orbit and the interior of the stars.

    Apart from a few first efforts that have been made to explain the origin of measured stellar-mass black hole spins, which were often hampered by oversimplifying assumptions, no systematic theoretical effort has been made yet towards this direction. Here we undertake the first detailed and systematic study of the origin of stellar-mass black hole spins. In this research project, we study all the physical processes taking place in the evolution of a binary star, progenitor of X-ray binaries, that can potentially affect the spin of the resulting black hole. Namely, we propose to study (i) the long-term stable accretion phase onto the black hole in low-mass X-ray binaries, (ii) the common envelope phase, (iii) the mass-transfer phase early in the life of a massive binary, and (iv) the tidal interactions of the black hole progenitor with the companion star in the post common envelope or post mass-transfer detached phase.

    All the physical processes mentioned above are modeled using detailed stellar structure and evolution, and binary mass-transfer calculations. Extensive grids of evolutionary tracks are created for each of the four main evolutionary phases, covering efficiently the parameter space of initial binary properties. The effects of stellar metallicity, which is an important factor in the evolution of massive stars, will be examined, including the limit of metal-free population III binaries. Moreover, at each step of our study, we will utilize the wealth of observational information available about the black hole X-ray binaries with measured black hole spin, making sure that our models produce systems consistent with the observed X-ray binaries. Effectively, we will uncover the evolutionary history of each of the black hole X-ray binaries with measured black hole spin, focusing on the processes that can affect the black hole spin.

  • The Cosmological Evolution of X-ray Binary Populations

    The Cosmological Evolution of X-ray Binary Populations

    A large scale population synthesis study, that models the X-ray binary populations from the first galaxies of the Universe until today. These models gave predictions about the global scaling of emission from X-ray binary populations with properties such as star-formation rate and stellar mass, and the evolution of these relations with redshift. Moreover, the energy feedback of X-ray binary populations in the intergalactic medium of the early universe is examined.

    One of my latest endeavours was a large scale population synthesis study, that models the X-ray binary populations from the first galaxies of the Universe until today (Fragos et al. 2013a). I used as input to our modeling the Millennium II Cosmological Simulation (Boylan-Kolchin et al. 2009) and the updated semi-analytic galaxy catalog by Guo et al. (2011) to self-consistently account for the star formation history and metallicity evolution of the Universe. My models, which were constrained by the observed X-ray properties of local galaxies , gave predictions about the global scaling of emission from X-ray binary populations with properties such as star-formation rate and stellar mass, and the evolution of these relations with redshift. My simulations showed that the X-ray luminosity density (X-ray luminosity per unit volume) from X-ray binaries in our Universe today is dominated by low-mass X-ray binaries, and it is only at z ∼ 2.5 that high-mass X_ray binaries become dominant. I also found that there is a delay of ∼ 1.1 Gyr between the peak of X-ray emissivity from low-mass X-ray binaries (at z ∼ 2.1) and the peak of star-formation rate density (at z ∼ 3.1). The peak of the X-ray luminosity from high-mass X-ray binaries (at z ∼ 3.9), happened ∼ 0.8 Gyr before the peak of the star-formation rate density, which is due to the metallicity evolution of the Universe.

    Although these models have only been constrained to observations of the local Universe, they have been shown to be in excellent agreement with X-ray observations of high redshift normal galaxies. Basu-Zych et al. (2013) studied the X-ray emmission from Lyman-break galaxies in the 4 Ms CDF-S field. These are actively star-forming galaxies at redshifts up to z ∼ 4, which probe the evolution of the HMXB population. Hornschemeier et al. (2013), using X-ray stacking techniques, studied the X-ray emission from low-mass early-type galaxies in the same 4 Ms CDF-S field. The X-ray emissions in these galaxies is believed to be dominated by the low-mass X-ray binary population. My contribution in both of these works was to model the galaxy sample selection and other selection effects, so that an appropriate comparison can be made between the synthetic models and the observations. It is remarkable that without any model fine-tuning, these models which have only been constrained to observations of the local Universe, are in excellent agreement with observations of both star-forming and early-type high-redshift galaxies.

    In a project that a graduate student led, under my supervision, we used the same library of synthetic models, along with empirical prescriptions for hot gas emission, to calculate the integrated X-ray luminosity of each galaxy in the Guo et al. (2011) catalog and create galaxy X-ray luminosity functions for several redshift bins (Tremmel et al. 2012). Comparing with observations by Tzanavaris & Georgantopoulos (2008), we found that our models are able to reproduce the general shape and evolution of the observed X-ray luminosity function. Our models predicted that the normalization of the X-ray luminosity function increases with redshift out to z ∼ 3 and then begins to decrease, consistently with the star formation history of the Universe.

    In follow-up work, I studied the energy feedback of X-ray binary populations in the intergalactic medium (Fragos et al. 2013b). The more energetic X-ray photons, because of their long mean-free paths, can escape the galaxies where they are produced, and interact at long distances with the intergalactic medium. This could result in a smoother spatial distribution of ionized regions, and more importantly in an overall warmer intergalactic medium. I found that at redshift z>8, XRBs are dominating the X-ray emission over the active galactic nuclei. Finally, I provided analytic prescriptions for the inclusion of the energy feedback from XRBs in cosmological simulations.

  • X-ray Binary Population Synthesis Studies in the Local Universe

    X-ray Binary Population Synthesis Studies in the Local Universe

    A theoretical population synthesis models for the formation and evolution of the X-ray binary populations in the local Universe and the study of the properties of their X-ray luminosity function at different galactic environments.

    Motivated by deep extra-galactic Chandra survey observations (Brassington et al. 2008, 2009), I worked on theoretical population synthesis models for the formation and evolution of the low-mass X-ray binary populations in the two elliptical galaxies NGC 3379 and NGC 4278. These simulations were targeted at understanding the origin of the shape and normalization of the observed X-ray luminosity functions (XLFs), and the relative contribution of different sub-populations of low-mass X-ray binaries at different luminosity ranges (Fragos et al. 2008). In a follow up study, I proposed a physically motivated prescription for the modeling of transient neutron star low-mass X-ray binary properties, such as duty cycle, outburst duration and recurrence time (Fragos et al. 2009a). This prescription enabled the direct comparison of transient low-mass X-ray binary synthetic models to the Chandra X-ray surveys of the two elliptical galaxies. This comparison suggested that transient low-mass X-ray binaries are very rare in globular clusters, and thus the number of identified transient low-mass X-ray binaries may be used as a tracer of the relative contribution of field and globular cluster populations.

    More recently, I worked on population synthesis models of black-hole X-ray binaries in the Milky Way. This work was motivated by recent developments in observational techniques for the measurement of black-hole spin in black-hole X-ray binaries. The accuracy of these techniques depend on the black hole spin misalignment with respect to the orbital plane, which can occur during the supernova explosion that forms the black hole. In my study, I presented synthetic models of Galactic black-hole X-ray binaries, and examined the distribution of misalignment angles, and its dependence on model parameters. This was the first study that provided robust theoretical constraints on the black hole spin tilts in black-hole X-ray binaries (Fragos et al. 2010).

    During the last 2 years, I have participated in a number of observational works, as the “in-house” theorist contributing to the data interpretation, and theoretical studies targeting at better understanding of the properties of X-ray binaries populations in local early-type galaxies (Brassington et al. 2012; Luo et al. 2012), in local star-forming galaxies (Tzanavaris et al. 2012), and in dense stellar environments such as globular clusters (Kim et al. 2009, 2013; Ivanova et al. 2012a; Strader et al. 2012; Roberts et al. 2012; Luo et al. 2013).

  • Reconstructing the Evolutionary History of Known X-ray Binaries

    Reconstructing the Evolutionary History of Known X-ray Binaries

    An innovative analysis method that allows the reconstruction of the full evolutionary history of known black-hole X-ray binaries back to the time of black hole formation formation. This analysis takes into account all the available observationally determined properties of a system, and models in detail four of its evolutionary evolutionary phases.

    I have applied an innovative analysis method that allows the reconstruction of the full evolutionary history of known black-hole X-ray binaries back to the time of black-hole formation. This analysis takes into account all the available observationally determined properties of a system, and models in detail four of its evolutionary evolutionary phases: mass transfer through the ongoing X-ray phase, tidal evolution before the onset of Roche-lobe overflow, motion through the Galactic potential after the formation of the BH, and binary orbital dynamics due to explosive mass loss and a possible black-hole natal kick at the time of core collapse. My results provide the most robust constraints on black-hole kicks due to asymmetries in the core collapse event. I applied this analysis on the Galactic black-hole X-ray binary XTE J1118+480, finding that right after BH formation, the system consists of a ≃ 6.0 − 10.0 M⊙ black hole and a ≃ 1.0 − 1.6 M⊙ main- sequence star, and that a large asymmetric natal kick (> 80km/s) is not only plausible but required for the formation of this system (Fragos et al. 2009b).

    Furthermore, I participated in four similar studies of the evolutionary history of known black-hole X-ray binaries: (i) the ultra-luminous X-ray source in the nearby elliptical galaxy NGC 3379, where we constrained the donor mass and orbital period at the onset of mass transfer within 1.15 − 1.4 M⊙ and 12.5 − 17 hr respectively (Fabbiano et al. 2006), (ii) the high mass X-ray binary M33 X-7, where we showed that system must have started its evolutionary path hosting a 94 − 115 M⊙ primary and a 32 − 38 M⊙ companion star in an orbit of 2.8-3.8 days (Valsecchi et al. 2010), (iii) the Galactic high mass X-ray binary Cyg X-1, where we found that the mass of the BH immediate progenitor is 15.0 − 20.0 M⊙ and that the BH has potentially received a small kick of Vkick ≤ 77 km s−1, and (iv) the nearby extragalactic high-mass X-ray binary IC10 X-1, where we found that the mass of the black-hole immediate progenitor is >31 M  and the magnitude of the natal kick imparted to the black hole is constrained to be <130 km/s.