Abstract: | The field of gravitational-wave astronomy began with the detection of a merging heavy stellar-mass black-hole binary. Many such compact binaries have been detected since then and more massive black-hole mergers are expected to be detected as the rest of the gravitational-wave spectrum is opened up in the coming years and decades. Gravity itself, however, is insufficient for bringing widely separated binaries to close-enough separations such that gravitational waves then drive inspiral towards merger. Compact binary formation thus requires various astrophysical mechanisms/channels that can extract enough orbital energy and angular momentum, many of which are wrought with uncertainty.
The main theme of this dissertation is the interplay of relativistic gravity and astrophysics, where gravitational radiation in the presence of binary interactions or astrophysical environments may provide new insights into how compact binaries form in our Universe.
We first extend the analysis of Peters and Mathews (1963) to binary systems whose masses may vary with time.
We consider the simplest case of isotropic mass loss, where we derive phase-dependent quadrupole-moment expressions. These new expressions contain contributions from the time-varying masses, which will thus also contribute to the gravitational radiation. We find that gravitational radiation from the changing binary mass can be orders of magnitude greater than that of the orbital motion if the mass loss occurs on a timescale much shorter than the orbital period. In the limit of zero mass variation, we recover the Peters and Mathews (1963) expressions. This work lays the foundation for the gravitational waveform source modeling of astrophysical binary systems that interact or are in the presence of astrophysical environments.
We then consider the case of supernova explosions in close binaries. Such explosions may be asymmetric such that a kick is imparted to the compact remnant, thus also kicking the binary into a new orbital configuration. We model the transition of close binaries from their pre-supernova state to their post-supernova state with an equation of motion that incorporates finite timescales for the supernova mass loss and natal kick. With integrated solutions to the equations of motion, we post-process the gravitational-wave emission to estimate at what level radiation reaction may be significant and prospects for detectability.
Building on the two previous works, we then model interacting binaries with an equation of motion that self-consistently treats gravitational radiation reaction. We apply this model to a few of astrophysical systems, including black-hole helium mergers, supernova mass loss and natal kicks in close binaries, and cold chaotic accretion onto supermassive black-hole binaries. Such systems have interaction/environmental timescales that can be comparable to less than the binary orbital period.
Here, we find that radiation reaction is mostly significant for supernova kicks in close binaries, while it can be neglected for the other two cases we considered here. The GW strain waveform, however, can be modified significantly depending on the strength of the mass variation and environmental drag.
We then turn our attention to common-envelope evolution, which is one of most important phases for the isolated binary channel as well as the most uncertain. We present a possible observable for this system: gravitational waves from accreting neutron stars that inspiral within the envelope. As covered in the earlier chapters of this dissertation, time-varying mass quadrupole moments source gravitational waves at leading order. As the neutron star accretes from the envelope, electron captures onto the surface combined with temperature gradients may induce density variations and thus form mountains on the surface. If such mountains are unaligned with the rotational axis of the NS, then a time-varying quadrupole moment is produced, thus sourcing gravitational-wave emission. The relatively hot temperatures associated with the hyper-Eddington accretion of the NS may also cause the crust to melt. If the melting timescale is comparable or longer than the common-envelope inspiral timescale, then such a system might be a gravitational-wave source of interest. We run several models accounting for hyper-Eddington accretion, neutron-star spin evolution, crust melting, and the uncertain common-envelope efficiency parameter. With these models, we estimate detectability prospects and discuss what new insights might be learned from such an observation.
To close this dissertation, we turn our attention for pulsar timing arrays and supermassive black-hole binaries. One of the astrophysics science cases for nanohertz gravitational-wave astronomy is to determine how astrophysical environments affect supermassive black-hole binaries in their evolution towards the gravitational-wave dominated regime.
Recently, electromagnetic transient surveys have been finding a growing catalog of quasars with year-like periodicity in their light curves. One possible explanation for the periodicity is the orbital motion of a supermassive black-hole binary. While a detection of the nanohertz stochastic gravitational-wave background has not yet been detected, upper limits are still astrophysically relevant and can constrain this growing catalog. We apply PTA upper limits to periodic blazars observed from the \textit{Fermi Gamma Ray Space Telescope} and constrain the binary fraction to $\lesssim 0.1\%$. Such constraints imply that the binary hypothesis alone is insufficient to explain the observed periodicity of blazars. |