Featured Research

Kruskal-Schwarzschild Instability (KSI) in a Strongly Magnetized Striped Wind

KSI simulation
Movie: Development of the instability from the linear to the non-linear phase and dripping of hot plasma from the current layer. The unmagnetized hot current layer is surrounded by cold strongly magnetized fluid with anti-aligned magnetic field lines going into and out of the page in, respectively, the top and bottom layers.

We study the linear and non-linear development of the Kruskal-Schwarzchild Instability in a relativisitically expanding striped wind. This instability is the generalization of Rayleigh-Taylor instability in the presence of a magnetic field. It has been suggested to produce a self-sustained acceleration mechanism in strongly magnetized outflows found in active galactic nuclei, gamma-ray bursts, and micro-quasars. The instability leads to magnetic reconnection, but in contrast with steady-state Sweet-Parker reconnection, the dissipation rate is not limited by the current layer`s small aspect ratio. We performed two-dimensional (2D) relativistic magneto-hydrodynamic (RMHD) simulations featuring two cold and highly magnetized ($1\leq\sigma\leq10^{3}$) plasma layers with an anti-parallel magnetic field separated by a thin layer of relativistically hot plasma with a local effective gravity induced by the outflow`s acceleration. Our simulations show how the heavier relativistically hot plasma in the reconnecting layer drips out and allows oppositely oriented magnetic field lines to reconnect. The instability`s growth rate in the linear regime matches the predictions of linear stability analysis. We find turbulence rather than an ordered bulk flow near the reconnection region, with turbulent velocities up to $\sim0.1$c, largely independent of model parameters. However, the magnetic energy dissipation rate is found to be much slower, corresponding to an effective ordered bulk velocity inflow into the reconnection region $v_{\rm in}=\beta_{\rm in}c$, of $10^{-3}\lesssim\beta_{\rm in}\lesssim 5\times10^{-3}$. This occurs due to the slow evacuation of hot plasma from the current layer, largely because of the Kelvin-Helmholtz instability experienced by the dripping plasma.

See my latest article on the KSI: Gill, R., Granot, J., & Lyubarsky, Y. 2017, Submitted to MNRAS

Gamma-ray bursts

Long duration GRBs are the most energetic explosions in the Universe where an isotropic equivalent of $E_{\rm iso}\simeq10^{49 - 54}$ ergs of energy is unleashed in gamma-rays in $~10$ sec. The spectrum is non-thermal and generally described by two power laws ($N_\gamma(E)\propto E^{-\alpha}$) with slopes $\alpha_{\rm low}\sim-1$ and $\alpha_{\rm high}\sim-2.3$ above and below the characteristic $\nu F_\nu$ peak energy $E_{\rm peak}\simeq250$ keV. The origin of this spectrum remains unclear and the answer lies in understanding the properties of the relativistic jet that is powered by the central engine, either a magnetar or a black hole.

GRB spectrum
Formation of GRB spectrum using my one-zone time-dependent kinetic code. Here, I start by injecting a quasi-thermal soft seed spectrum (Wien peak and flat $F_\nu$), and let it evolve in time while accounting for Compton scattering, pair production/annihilation, Coulomb interaction and volumetric heating of pairs.

My recent study with Chris Thompson at CITA and two other works co-authored by me present a complete description of the physics that gives rise to the GRB spectrum. We show that the low energy spectrum forms pre-breakout when the jet, powered by a central black hole in our model, is mildly relativistic and working its way out of the confining medium provided by the Wolf-Rayet progenitor. The inertia in the jet is dominated by the magnetic field and the low-energy spectrum is formed by quasi-thermal comptonization in a photon-rich $e^-e^+$-pair fireball, with no baryon contamination. The high-energy spectrum forms post-breakout when the jet is relativistic and optically thin. The dissipation of magnetic energy is provided by the interaction of the relativistic magnetofluid with baryons entrained in the jet from the confining medium. In both the optically thick and thin phases, the pairs are heated volumetrically due to the damping of hydromagnetic turbulence in the flow. To simulate the high energy spectrum, I developed a one-zone kinetic code with all radiative processes treated in an exact manner.

We have developed a cogent explanation for the prompt emission mechanism in a series of articles:


The soft $\gamma$-ray repeaters (SGRs) dissipate large amounts of magnetic energy in the form of highly luminous gamma-ray bursts called giant flares. How the flares are triggered and what controls their duration and energy is still an open question. The energetics of the bursts, however, do point towards the extremely strong fields that these objects are endowed with. According to the TD95 model of magnetar bursts, the giant flares involve major restructuring of the global magnetic field of the star on timescales longer than the Alfven crossing time of the magnetosphere ($t_A\sim10^{-5}$ s). This takes place in concert with a magnetic reconnection event in the magnetosphere that ultimately triggers the hyperflare.

Magnetospheric Reconnection
This figure displays the setup of the different reconnecting current layers. The macroscopic Sweet-Parker layer with length $L \sim 10^5$ cm and width $\delta \sim 0.01$ cm is the largest of the three. This layer is then thinned down vertically as strong magnetic flux is convected into the dissipation region. The Hall reconnection layer, represented by the dark gray region, develops when $\delta$ becomes comparable to the ion-inertial length $d_i$ . The system makes a transition from the slow to the impulsive reconnection and powers the main flare. The tiny region embedded inside the Sweet-Parker layer is the super-hot turbulent current layer, which aids in creating sufficient anomalous resistivity to facilitate the formation of the Sweet-Parker layer. The strongly accelerated plasma downstream of the reconnection layer is trapped inside magnetic flux lines and forms a plasmoid moving at some speed $V$ . This plasmoid is then finally ejected during the initial spike when the external field undergoes a sudden relaxation

One interesting observation these models do not resolve is the connection between the precursor burst and the main flare. In two out of the total three giant flares that have been observed, a precursor burst of very short duration was detected a few seconds before the main flare. In my study, I show that the precursor bursts are not only causally connected to the main flares but also triggers them. I examine two reconnection models that explain how the wound up flux tubes inside the NS and the sheared field lines in the magnetosphere can form tangential discontinuities and dissipate stored magnetic energy. In this work, I introduced a novel idea of Hall reconnection operating in the inner magnetosphere which ultimately leads to an explosive energy release.

Work on magnetars:


Unlike weak interactions, that are responsible for the radioactive decay of subatomic particles, strong interactions have thus far been experimentally verified to preserve $CP$-symmetry. However, this observation is in conflict with QCD which necessarily violates CP. This can be remedied in many ways but the one most favored solution to the Strong CP problem has been the introduction of a new particle into the Standard Model, dubbed the axion. In several extensions of the Standard Model axion-like particles (ALPs) are produced naturally. Such particles remain elusive due to their very small mass ($10^{-6} < m_a < 10^{-2}$ eV for the QCD axion) and extremely weak coupling to matter and radiation with $g_{a\gamma\gamma} \lesssim 10^{-10}\mbox{ GeV}^{-1}$. The limit on $g_{a\gamma\gamma}$, derived from astrophysical arguments, in the given mass range is a few orders of magnitude larger than that predicted by theoretical axion models. Several laboratory experiments have been conducted and many are currently underway to exclude as much of the parameter space as possible.

Exclusion Plot

In my work on ALPs, I take a novel approach to constrain the properties of ALPs. I use the polarized light of mWDs to determine the absolute upper limit on $g_{a\gamma\gamma}$ as a function of $m_a$. My treatment is based on the observation that light is polarized as it traverses the magnetosphere encompassing the tenuous plasma permeated by the $\sim$ 100 MG magnetic field of some mWDs. Since the photon-axion interaction in a strongly magnetized plasma leads to additional polarization, the observed degree of polarization can be used to constrain axion properties if the plasma contribution is correctly modeled. The derived constraints in my work are hitherto the best estimates of ALP parameters to come from compact objects. Moreover, the results of this study exceed estimates from laboratory experiments for $10^{-6}\lesssim m_a\lesssim10^{-5}$ eV.

Work on Axions: