Fully Kinetic Shearing-box Simulations of Magnetorotational Turbulence in 2D and 3D. I. Pair Plasmas

Abstract

The magnetorotational instability (MRI) is a fundamental mechanism determining the macroscopic dynamics of astrophysical accretion disks. In collisionless accretion flows around supermassive black holes, MRI-driven plasma turbulence cascading to microscopic (i.e., kinetic) scales can result in enhanced angular-momentum transport and redistribution, nonthermal particle acceleration, and a two-temperature state where electrons and ions are heated unequally. However, this microscopic physics cannot be captured with standard magnetohydrodynamic (MHD) approaches typically employed to study the MRI. In this work, we explore the nonlinear development of MRI turbulence in a pair plasma, employing fully kinetic particle-in-cell (PIC) simulations in two and three dimensions. First, we thoroughly study the axisymmetric MRI with 2D simulations, explaining how and why the 2D geometry produces results that differ substantially from 3D MHD expectations. We then perform the largest (to date) 3D simulations, for which we employ a novel shearing-box approach, demonstrating that 3D PIC models can reproduce the mesoscale (i.e., MHD) MRI dynamics in sufficiently large runs. With our fully kinetic simulations, we are able to describe the nonthermal particle acceleration and angular-momentum transport driven by the collisionless MRI. Since these microscopic processes ultimately lead to the emission of potentially measurable radiation in accreting plasmas, our work is of prime importance to understand current and future observations from first principles, beyond the limitations imposed by fluid (MHD) models. While in this first study we focus on pair plasmas for simplicity, our results represent an essential step toward designing more realistic electron–ion simulations, on which we will focus in future work.

Sub-GeV Gamma Rays from Nearby Seyfert Galaxies and Implications for Coronal Neutrino Emission

“…Alternatively, magnetic reconnections accelerate nonthermal particles efficiently, especially in the relativistic regime (σ mag = B 2 /(4πnmc 2 ) > 1; Hoshino & Lyubarsky 2012;Guo et al 2020). PIC simulations have demonstrated that the development of MRI turbulence triggers magnetic reconnections, leading to nonthermal particle production (e.g., Hoshino 2015; Kunz et al 2016;Bacchini et al 2022). Kheirandish et al (2021) suggested such a magnetic reconnection model to explain the neutrino data of NGC 1068 (see also Fiorillo et al 2023;Mbarek et al 2023).…”

Balanced Turbulence and the Helicity Barrier in Black Hole Accretion

“…Several mechanisms for energy dissipation have been proposed in the context of turbulent cascades, including cyclotron heating (Kennel & Engelmann 1966;Isenberg 2004;Isenberg & Vasquez 2007, 2011Arzamasskiy et al 2019;Klein et al 2020), stochastic heating (McChesney et al 1987Chandran et al 2010;Arzamasskiy et al 2019;Cerri et al 2021), Landau damping (Howes 2010;Kawazura et al 2020;Arzamasskiy et al 2023), as well as reconnection and Fermitype acceleration in relativistic plasmas (Comisso & Sironi 2018;Zhdankin et al 2018;Comisso & Sironi 2019;Zhdankin et al 2019Zhdankin et al , 2021Bacchini et al 2022).…”

RelSIM: A Relativistic Semi-implicit Method for Particle-in-cell Simulations

“…In general, we compare the results obtained with the RelSIM to those obtained with the RelIMM as well as with a standard explicit-PIC code. For the explicit runs, we employ ZELTRON, which is a state-ofthe-art tool utilized for many production applications in relativistic astrophysics (e.g., Cerutti et al 2013;Zhdankin et al 2017;Werner et al 2018;Parfrey et al 2019;Mehlhaff et al 2021;Bacchini et al 2022a;Galishnikova et al 2023). ZELTRON uses a standard explicit-leapfrog discretization for particle and field equations and a numerical grid based on a Yee lattice.…”

“…Several codes employed in astrophysical and laboratory plasma research, e.g., EPOCH (Arber et al 2015), OSIRIS, (Fonseca et al 2002), SHARP (Shalaby et al 2017), SMILEI (Derouillat et al 2018), TRISTAN V2 (Hakobyan et al 2023b), VORPAL (Nieter & Cary 2004), VPIC (Bird et al 2022), WARPX (Fedeli et al 2022), ZELTRON (Cerutti et al 2013;Bacchini et al 2022a), etc. 3 have been employing the explicit-PIC approach for a long time, and they have successfully attacked many open problems in relativistic astrophysics (e.g., Spitkovsky 2008;Zhdankin et al 2017;Comisso & Sironi 2018;Guo et al 2021;Sironi 2022), even including quantum electrodynamics and strong-gravity effects (e.g., Parfrey et al 2019;Crinquand et al 2020;Sridhar et al 2021;El Mellah et al 2022;Hakobyan et al 2023a;Galishnikova et al 2023;Grošelj et al 2023), or frame transformations appropriate, e.g., for expanding/shearing plasmas (Riquelme et al 2012;Hoshino 2015;Sironi & Narayan 2015;Bacchini et al 2022a;Tran et al 2023). The simplicity of explicit PIC also allows for efficient implementation on new architectures such as graphicsprocessing units; see, e.g., PICONGPU (Burau et al 2010) and ENTITY (H. Hakobyan et al 2023, in preparation).…”