General Relativity & Quantum Cosmology

In this talk, I will explore key open questions in our understanding of neutron star mergers and their multi-messenger emission, highlighting recent progress made by our group. I will then introduce some of the first applications of AthenaK to compact binary mergers and outline our plans for its future development.

With the ongoing transition toward exascale computing to tackle a range of open questions via numerical simulations, the development of GPU-optimized codes has become essential. In this talk, I will highlight the key features of AsterX, a novel open-source, modular, GPU-accelerated general relativistic magnetohydrodynamic (GRMHD) code for fully dynamical spacetimes in 3D Cartesian coordinates. Built for exascale applications, AsterX integrates with CarpetX, the new driver for the Einstein Toolkit, leveraging AMReX for block-structured adaptive mesh refinement (AMR). The code employs the flux-conservative Valencia formulation for GRMHD, and uses high-resolution shock capturing schemes to ensure accurate hydrodynamic modeling. Alongside discussions on the ongoing code development, I will also present the results of comprehensive 1D, 2D, and 3D GRMHD tests conducted on OLCF's Frontier supercomputer, highlighting AsterX's performance gains through subcycling in time and demonstrating its scaling efficiency across thousands of nodes.

In certain scenarios, the accreted angular momentum of plasma onto a black hole could be low; however, how the accretion dynamics depends on the angular momentum content of the plasma is still not fully understood. We present three-dimensional, general relativistic magnetohydrodynamic simulations of low angular momentum accretion flows around rapidly spinning black holes (with spin $a = +0.9$). The initial condition is a Fishbone-Moncrief (FM) torus threaded by a large amount of poloidal magnetic flux, where the angular velocity is a fraction $f$ of the standard value. For $f = 0$, the accretion flow becomes magnetically arrested and launches relativistic jets but only for a very short duration. After that, free-falling plasma breaks through the magnetic barrier, loading the jet with mass and destroying the jet-disk structure. Meanwhile, magnetic flux is lost via giant, asymmetrical magnetic bubbles that float away from the black hole. The accretion then exits the magnetically arrested state. For $f = 0.1$, the dimensionless magnetic flux threading the black hole oscillates quasi-periodically. The jet-disk structure shows concurrent revival and destruction while the gas efficiency at the event horizon changes accordingly. For $f \geq 0.3$, we find that the dynamical behavior of the system starts to approach that of a standard accreting FM torus. Our results thus suggest that the accreted angular momentum is an important parameter that governs the maintenance of a magnetically arrested flow and launching of relativistic jets around black holes.

Observations of luminous black holes in X-ray binaries and Seyfert galaxies show power-law emission, thought to originate from photons that inverse Compton scatter off a hot electron cloud. If the coronal electrons are heated by magnetic dissipation, i.e. reconnection or turbulence, then one might expect to observe direct synchrotron emission in the radio/mm from these electrons. However, because timing studies constrain the X-ray emission to be within ~10 rg of the central black hole, the direct synchrotron emission from this compact volume would be strongly self-absorbed until much further away from BH. In this talk, I will question the de facto definition of the corona as a compact, X-ray-emitting region and shift instead to a paradigm where the corona encompasses multiple layers with distinct spectral components. Motivated by highly-magnetized winds found in GRMHD simulations, I will present a model for such an extended, outflowing corona. I will discuss this model in the context of radio-quiet AGN, where recent observations have demonstrated the presence of compact mm emission.

Tilted magnetically arrested disks (MADs) around black holes—where strong magnetic fields regulate accretion and jets—exhibit striking alignment behavior dictated by black hole spin direction. Using 3D general-relativistic magnetohydrodynamic (GRMHD) simulations of tilted MADs, we find that prograde disks align via a two-stage process: an initial rapid alignment phase, ending at the flux saturation timescale, followed by a slower, spin-independent phase. In contrast, retrograde MADs remain persistently misaligned, with their inner disks precessing four times faster than weakly magnetized systems—a potential explanation for high-frequency quasi-periodic oscillations (QPOs). By analyzing magnetic and hydrodynamic torques within ideal GRMHD, we show that alignment in prograde disks is dominated by electromagnetic stresses from the magnetosphere. However, the same magnetic forces— which always act to align the disk with the black hole spin—are significantly weaker in retrograde disks, allowing opposing hydrodynamic torques to dominate. These results suggest that jets alone may not be sufficient to align MAD disks, instead highlighting the magnetosphere’s crucial role in mediating spin-disk coupling.

What powers the hard, non-thermal X-rays from accreting compact objects has been a longstanding mystery. In my talk, I will address the underlying question of what energizes the particles of the Comptonizing “corona” against the strong inverse Compton (IC) cooling losses with first-principle particle-in-cell simulations of magnetic reconnection subject to IC cooling in magnetically dominated electron-positron plasmas, and in mildly-magnetized electron-ion plasmas. I will also show---using results of global resistive GRMHD simulations of accreting black holes---that the black hole jet sheath is a site of efficient electromagnetic dissipation through processes such as magnetic reconnection and turbulence. The distribution of bulk motions of the radially outflowing plasma along the jet sheath also resembles a Maxwellian distribution with an effective bulk temperature of a few 100 keV, and this could be a candidate for the Comptonizing corona.

A new version of the Athena++ AMR MHD code implemented using the Kokkos library for performance-portability will be described. The code is being used for a variety of problems related to accretion onto (and merger of) compact objects. This talk will focus on results from general relativistic radiation MHD models of accretion in luminous systems such as X-ray binaries and AGN. The models allow observationally important quantities such as the radiative efficiency, variability, and kinetic feedback in winds to be measured over a range of luminosities, from sub-Eddington to highly super-Eddington. A new “cyclic-zoom” approach for modeling such flows over very long time scales will also be briefly introduced.

An alternative interpretation of the EHT observations of M87, involving a strongly magnetized ergomagnetosphere, that expels essentially of the gas supplied at large radius as a jet-collimating MHD wind is described. In this case, the emitting regions are approximated as rapidly evolving flares or sparks. Six generic consequences, that can be sought in existing data, are described.