I'm broadly interested in all kinds of astrophysical systems, from stars, planets to compact objects. Specifically, I have been involved in the following topics:
Stripped helium stars are core helium-burning stars whose envelopes were lost due to some interactions with a nearby companion. Throughout their lifetime, these stars could experience significant tidal spin-up. Observationally, stripped helium stars are found in binaries with massive Wolf-Rayet stars and (less massive) B-type subdwarf (sdB) stars. The former may collapse to a fast-rotating black hole to be detected as gravitational-wave signals, while the latter’s rotation rates can be measured with line-broadening or asteroseismology techniques. I studied the rotational evolution of these stars with tidally excited internal gravity waves. For Wolf-Rayet binaries, this process may produce fast (but not maximally) rotating black holes. For sdB stars, the results agree with the observed trend of tidal synchronization.
My paper on Wolf-Rayet tidal spin-up: ApJ, 952, 53
My paper on sdB tidal spin-up is coming soon.
The observational versus the modeling trend of sdB tidal synchronization.
Many planets are discovered in short-period orbits, where they excite stellar tides that transfer their orbital angular momentum to the star, causing their orbits to decay. A particularly interesting process is called tidal resonance locking. In this scenario, the orbit of the planet is locked with one of the evolving oscillation modes inside the star. Since the stellar evolution is almost never influenced by the planet, this process causes the planetary orbital decay to occur on the stellar main-sequence timescale. I studied this process for different stellar hosts and planets and found that resonance locking occurs for all small planets, or those massive planets around different types of hosts. This predicts that the distribution of exoplanets should be a strong function of their masses and their host star types.
My paper on exoplanet orbital decay with resonance locking: ApJ, 918, 16
When a planet is created, it will soon get trapped in one of the stellar oscillation modes and starts evolving with this mode (resonance locked), hence decays its orbit.
An open question in stellar astrophysics is the inconsistency between the overabundance of blue supergiants and their predicted short lifetime from classical stellar evolution theory. Several theoretical explanations have been proposed, which may be tested by their predictions for some characteristic photometry variability. I analyzed the light curves of 20 blue supergiants in the Large Magellanic Cloud obtained from the Transiting Exoplanet Survey Satellite (TESS) mission and found a characteristic signal in the low-frequency range for all the targets. The peak frequency of the signal suggests its possible relation with sub-surface convection caused by the iron opacity peak in the star, and the shape of the spectra might be explained by the propagation of high-order, damped gravity waves.
My paper preprint on blue supergiant variability: arXiv:2310.19546
The characteristic signal (periodogram) we found for TIC 425084139, on top of its GAIA star field.
The internal rotational dynamics of massive stars are poorly understood. Observations of low-mass stars suggest an efficient angular momentum transport mechanism should be at work, and such mechanisms could be caused by some hydrodynamic or magneto-hydrodynamic instabilities in the stellar interior. I studied the effects of the baroclinic instability and the magnetic Tayler instability in differentially rotating radiative zones. Although the baroclinic instability may occur, the Tayler instability is likely to be more efficient for angular momentum transport. I implemented Tayler torques as prescribed in Fuller et al .2019 into stellar evolution models of massive stars, finding that they remove the vast majority of the core's angular momentum as it contracts, between the main sequence and helium-burning phases of evolution.
My paper on angular momentum transport: MNRAS, 488, 4338
In the stellar evolution model, the majority (>99%) of core angular momentum is lost before the helium burning phase.
A major uncertainty in stellar theory is the angular momentum transport within the star that determines its core rotation rate and the resulting spins of the compact object remnants (neutron stars and black holes). By implementing an updated magnetic angular momentum transport prescription into models of high-mass stars, I studied the rotation rates of neutron stars and black holes born from single-star evolution. I found that these objects usually rotate slowly, with neutron star rotation periods of 50-200 ms, and typical black hole spins of 0.01. The results on black hole spins are possibly explaining the low effective χ of most black hole binaries detected by LIGO.
My paper on neutron star spins: MNRAS, 488, 4338
Our paper on black hole spins: ApJL, 881, L1
LIGO has measured low spins for most BH mergers via GW detections, which is consistent with our model.
Observations have found supermassive black holes in very early galaxies. People have argued that these black holes may form by growing from some less massive “seed” black holes. To do this these seeds need to sink to the densest region in their host galaxies, which may not be true as these galaxies are usually very chaotic and clumpy. I studied the dynamics of these seed black holes with state-of-the-art high-resolution cosmological simulations and found that seed black holes less massive than 100 million solar masses (all but the already-supermassive seeds) cannot efficiently sink in typical high-z galaxies. This points out a potential problem in the current formation models of supermassive black holes.
My paper on sinking of seed black holes: MNRAS, 508, 1973
High-z galaxies are clumpy and chaotic objects, which makes the dynamics of BHs inside extremely complicated.
When a moving body travels through a cloud of lighter particles, it will lose its momentum by having successive two-body encounters with the background particles. If the gravity from background particles is treated as if it's from a smooth continuum, it is hence necessary to add an extra term to the equation of motion known as the dynamical friction (DF). Historically, dynamical friction is usually evaluated by Chandrasekhar's dynamical friction formula, which only applies to an infinite, homogeneous and isotropic background with Maxwellian velocity distribution. I developed a new DF estimator which could be applied to an arbitrary phase space configuration. When applying to N-body simulations, my new estimator also avoids several ill-defined quantities in Chandrasekhar's formula and is more consistent and computationally efficient.
My paper on the dynamical friction estimator: MNRAS, 519, 5543
True 2-body scattering effectively extract a "dynamical friction" to the test particle compared to smooth gravity field.
Back to top