[論文レビュー] Proton Acceleration by Collisionless Shocks in Supermassive Black Hole Coronae: Implications for High-Energy Neutrinos

Minh Nhat Ly, Yoshiyuki Inoue+2 more|arXiv (Cornell University)|Jan 5, 2026

Astrophysics and Cosmic Phenomena被引用数 0

ひとこと要約

この研究は第一原理1D3V PICシミュレーションを用いて、SMBHコロナでの拡散ショック加速が陽子を効率的に加速し（ショック運動エネルギーの約10%に相当）、幅広いパラメータ範囲でHadronicニュートリノ生成をAGNコロナに対して支持することを示す。

ABSTRACT

Recent observations by the IceCube Neutrino Observatory have revealed a significant excess of high-energy neutrinos from nearby Seyfert galaxies, such as NGC~1068, without a corresponding flux of high-energy gamma-rays. This suggests that neutrinos are produced via hadronic interactions in a region opaque to gamma-rays, likely a hot corona surrounding the central supermassive black hole. However, the mechanism responsible for accelerating the parent protons to the required energies ($\sim 100$ TeV) remains an open question. In this study, we investigate diffusive shock acceleration (DSA) in active galactic nucleus (AGN) coronae using a suite of one-dimensional Particle-in-cell (PIC) simulations spanning a broad range of plasma parameters. We find that DSA is a robust and efficient mechanism for proton acceleration, consistently channeling approximately 10\% of the shock's kinetic energy into non-thermal ions, even for shocks with sonic Mach number as low as $ M_s \approx 2$. In contrast, the efficiency of electron acceleration is highly variable and less efficient ($<1\%$) in our parameter survey. These findings provide strong, first-principles support for the hadronic models of neutrino production in AGN and offer quantitative constraints that can explain the observed gamma-ray deficit.

研究の動機と目的

• Motivate hadronic neutrino production in AGN by identifying viable proton acceleration mechanisms in SMBH coronae.
• Quantify how collisionless quasi-parallel shocks in SMBH coronae accelerate protons across realistic parameter ranges (Mach numbers, temperatures, magnetic fields).
• Provide first-principles constraints on proton and electron acceleration efficiencies to explain gamma-ray deficits alongside IceCube neutrinos.

提案手法

• Perform large-scale 1D3V PIC simulations with the SMILEI code of quasi-parallel collisionless shocks.
• Survey a wide parameter space: sonic Mach number Ms, Alfvénic Mach number MA, shock velocity, and ion-to-electron temperature ratio Ti/Te.
• Use a reduced ion-to-electron mass ratio mi/me=100 with relativistic Maxwell-Jüttner distributions for initial temperatures.
• Initialize a reflective boundary to form a shock and analyze downstream non-thermal particle spectra and upstream magnetic turbulence.
• Identify dominant upstream instabilities (non-resonant Bell-type) and characterize the resulting magnetic field amplification and particle acceleration pathways

Figure 1: Schematic illustration of an accretion shock formed by the in-falling flow in the hot BH corona. Protons are accelerated to high energies and produce high-energy neutrinos through hadronic $pp$ and $p\gamma$ interactions. Neutrinos can escape from the system and be observed by IceCube, whi

実験結果

リサーチクエスチョン

• RQ1 Can diffusive shock acceleration robustly accelerate protons in SMBH corona conditions with trans-relativistic shocks (v_sh/c ~ 0.3) and low Ms?
• RQ2 What are the proton and electron acceleration efficiencies across a broad range of Ti/Te, Ms, MA, and v_pt, and how do these affect hadronic neutrino production in AGN coronae?
• RQ3 What is the energy partition between thermal and non-thermal particles, and how does magnetic turbulence develop upstream to sustain DSA?
• RQ4 Do the simulated acceleration rates depend strongly on shock parameters, and is there a universal scaling for the proton acceleration efficiency?

主な発見

• Protons are efficiently accelerated by DSA across a broad parameter range, consistently channeling about 10% of the shock’s kinetic energy into non-thermal ions.
• Electron acceleration is significantly less efficient (<1%) and more sensitive to shock conditions than proton acceleration.
• Proton energy gain follows a two-phase evolution, with linear-in-time growth after a transient early stage, consistent with DSA; the late-time normalized acceleration rate is roughly C^-1 ≈ 0.1 across runs.
• The non-thermal ion spectra have power-law tails with index si ≈ 2, and the electron spectra are softer, with indices depending on Ms and Ti/Te.
• Upstream turbulence is dominated by non-resonant (Bell-type) instabilities in most cases, enabling sustained scattering for repeated shock crossings; low MA cases show weaker turbulence and slightly reduced acceleration.
• The results provide first-principles support for hadronic-neutrino production in AGN coronae and help explain gamma-ray deficits observed with IceCube neutrinos (e.g., NGC 1068).

Figure 2: Snapshot of shock sctructures in fiducial model (run T10) at $t\approx 256~\Omega_{\rm ci}^{-1}$ . (a, b) The position-momentum ( $x-p_{x}$ ) phase space distributions for ions and electrons. (c) Ion density profile, with the shock position ( $x_{\rm sh}$ ) marked by the red dashed line. (