Many galaxies show strong star formation activity in their central regions. In galaxies that also host an active galactic nucleus (AGN), star formation may occur locally within the AGN accretion disk itself. Low-mass stars constitute the majority of the stellar population, and as they reach the end of their life cycles, they evolve into white dwarfs. We therefore expect to find a population of white dwarfs embedded in and comoving with the AGN accretion disk. The disk provides a dense environment in which white dwarfs can efficiently accrete gas and grow in mass. As their mass approaches the Chandrasekhar limit of 1.4 solar masses, the temperature becomes high enough to trigger runaway nuclear fusion, eventually leading to a Type Ia supernova. Type Ia supernovae are crucial both as standard candles for cosmology and as major contributors to the chemical enrichment of the universe. In this work, we investigate the population of white dwarfs in AGN disks, their mass growth through accretion, and their eventual collapse into Type Ia supernovae. We show that the accretion of white dwarfs in AGN disks is a viable channel for producing type 1a supernovae. We discuss the resulting consequences for the evolution of the AGN and its accretion disk, the chemical enrichment of galactic centers, and the universality of Type Ia supernovae as standard candles.

The quasar Eigenvector-1/Main Sequence (E1/MS) is a practical roadmap that organizes AGN spectral diversity by accretion state, letting us compare objects in a common, physically motivated trend. Past applications of the E1/MS sequence made it possible to identify the effects of intrinsic viewing angle, black hole mass, and luminosity. We propose an E1/MS-driven framework to clarify the accretion status of gamma-ray-emitting AGN with an emphasis on radio-loud narrow-line Seyfert 1s (NLSy1s) and ​Population A sources. We argue that the E1/MS context is ideal for flagging super-Eddington candidates among radio-loud NLSy1s and other Population A sources that also host relativistic jets. An optimal strategy would be to place gamma-ray AGN on E1 using optical/UV spectroscopy, to derive disk-based bolometric luminosities from line or IR reprocessing (or jet-quiet SED states), and to correct single-epoch black hole masses for orientation and radiation pressure biases. This strategy should identify super-Eddington candidates and reflect a physical link between accretion state and gamma-ray efficiency: dense broad-line region/torus photon fields in high-Eddington-ratio systems should boost the external Compton emission, whereas mass loading and radiative drag shape moderate bulk Lorentz factors in gamma-ray emitters observed with VLBI. We apply the method based on the E1/MS criteria to targets for which high-S/N optical and UV spectra as well as multi-frequency coverage are available.

Narrow-Line Seyfert 1 (NLS1) galaxies exhibit narrow permitted Hβ lines (FWHM < 2000 km/s) and strong Fe II multiplets with low black hole (BH) masses (< 10⁸M⊙) and high Eddington ratios. The BH masses and predominantly disk-like hosts suggest that NLS1s are early-stage Active Galactic Nuclei (AGN). They host the lowest-mass super-massive BHs that can launch relativistic jets and offer insights into the conditions necessary for jet formation. Available optical spectral catalogues have low S/N ratios, and classifying NLS1s using them is challenging. Using FORS2 observations of ~100 candidate NLS1s, we aim to classify the sources and perform detailed spectral modelling of Hβ, [OIII], and the Fe II pseudo-continuum using Gaussian, Lorentzian and Voigt profiles. Based on preliminary results, we found 16 changing-look AGN and a contamination fraction of ~28% in our data. We study how, in NLS1s, the broad Hβ line profile transitions from a Lorentzian to a Keplerian motion-dominated Gaussian with increasing BH mass, to determine whether this change is an essential part of intraclass evolution. By modelling the lines with Voigt profiles, we test whether the sources previously best described by Lorentzian or Gaussian functions are dominated by the turbulent or Keplerian component, respectively. In this talk, I will present the spectral analysis of NLS1s and changing-look AGN and discuss our findings and future plans.

Narrow-line Seyfert 1 galaxies (NLS1s) are a class of active galactic nuclei (AGN) and were long believed to be radio-silent and incapable of producing relativistic jets; however, recent studies have demonstrated otherwise. The Fermi Large Area Telescope (LAT) made a groundbreaking discovery by detecting gamma-ray emission from the NLS1 source PMN J0948+0022, providing clear evidence that some of these sources can host relativistic jets in this class of AGN. This challenged the idea we had of NLS1s and their diversity.

Over the years, many new sources have been discovered, and my researchers have focused on analyzing these findings by systematically searching for additional gamma-ray-emitting NLS1s using data from the Fermi-LAT. For this, a series of analyses are run to evaluate the test statistic (TS) value, which provides a measure of the significance of potential detection.

This poster will present an overview of narrow-line Seyfert 1 galaxies (NLS1s), the methodology applied in this study, relevant previous findings, the analysis procedure, and the automation of code used to conduct the analysis. Current results will be discussed, highlighting the test statistic (TS) values obtained thus far. As this project represents ongoing research, additional TS values will be identified over time, culminating in a more comprehensive analysis.

Cosmic rays (CRs) are accelerated in astrophysical systems, such as starburst galaxies, active galactic nuclei and large-scale shocks. These accelerators are usually embedded in the overdense part of the cosmic web, i.e., galaxy clusters and filaments, which are threaded with magnetic fields. Many studies only model the the transport of charged particles in magnetic fields and ignore the secondary neutrons. We explored an alternative route for CR escape—assisted by temporary charge neutral particles, i.e., neutrons. These neutrons are produced in the hadronic interactions of CR protons, and with the aid of time dilation, they may traverse distances larger than galactic scales before decaying into protons. We capture this alternative CR escape channel with a stochastic Poisson term in the transport equation, which also accounts for other processes that attenuate CR flux. We model the CR escape with the neutron-assisted channel for several test cases, representing host sites being a galaxy, galaxy cluster, and filament. We find that the neutron-assisted escape channel is most efficient for ultra-high-energy cosmic rays (E >~ 1e18 eV) and may modify the energy spectrum of CRs at the highest energies. I will present our methodology for modelling CR escape and highlight the importance of incorporating the physics of neutron-assisted CR escape.

The ultra-relativistic, highly collimated jets generated by Gamma-Ray Bursts (GRBs) provide crucial insights into particle emission. These jets also reveal the physical mechanisms driving the rapid release of high-energy gamma-ray photons. We discuss time-resolved spectroscopy and flux variability for the ultra-long GRB 220627A. The analysis spans a duration exceeding 1200 seconds using Fermi telescope data. Two prompt emission episodes observed by Fermi-GBM, separated by more than 500 s, were analyzed. Due to its unique characteristics, GRB 220627A serves as an excellent source for studying particle emission processes, small-scale variability, and the properties of its central engine. To investigate gravitational lensing, the time bins of the first episode were correlated with those of the second episode. A coherent relationship was observed between flux and photon spectral distribution. This relationship was modeled using an exponentially cut-off power law model for both episodes. The MeV-to-GeV photon ratio detected by LAT is inconsistent between the two episodes. High-energy gamma-ray photons were only detected by LAT for up to 700 seconds, which further rules out gravitational lensing but suggests that the progenitor underwent a burst into an ultra-long GRB with two episodes. Our findings from spectral analysis reveal characteristics most consistent with those of an ultra-long GRB. Parameters such as isotropic energy, spectral signatures, and burst duration align with the established limits for a blue supergiant progenitor, as described in the literature.

Our understanding of jet kinematics in high-redshift (z≥3) quasars is still rather limited, based on a sample of less than about fifty objects. It is difficult to perform such measurements for multiple reasons. In this work, we present very long baseline interferometry (VLBI) observations of the surprisingly rich radio jet structure of the powerful blazar J1429+5406 at z=3.015, observed at five different frequencies (1.7–15 GHz) between 1994 and 2018. While the outer jet components at ∼20–40 milliarcsecond (mas) angular separation from the core show no apparent proper motion, three components within 10 mas distance exhibit significant proper motions of (0.045–0.16) mas year^‑1, including one that is among the fastest-moving jet components at z≥3 known to date. The core brightness temperature well exceeds the equipartition limit, indicating Doppler-boosted radio emission. Based on the proper motion of the innermost component, we derive a small jet inclination with respect to the line of sight (within about 5°), confirming the blazar nature of the source. Recently, we analysed new high-sensitivity VLBI measurements of this blazar taken by the Very Long Baseline Array in 2024. At 1.5 GHz, the radio map clearly reveals a complex, extended structure around the source, reaching up to ~400 mas, rarely, if ever, seen in a high-redshift blazar. Here, we discuss possible explanations of this peculiar radio structure.

Glitches in neutron stars originate from the sudden transfer of angular momentum between superfluid components and the observable crust.
By modeling this glitch dynamics—including vortex motion, mutual friction, and angular momentum exchange—we can probe the dense matter equation of state. We match theoretical predictions of glitch rise times, overshoot patterns, and relaxation timescales to the well-documented observations of the 2016 Vela glitch. Our model incorporates microphysical parameters such as the mutual friction coefficient \(\mathcal{B}\), which in the core arises from electron scattering off magnetized vortices and in the crust from Kelvin wave excitation during vortex–lattice interactions. Our Markov Chain Monte Carlo analysis of the timing residuals reveals detailed glitch dynamics: the crustal superfluid couples on timescales of $\sim100$ seconds, the core exhibits overshoot behavior due to strong central behavior, and the inner crust shows weak entrainment, with $\sim70\%$ of free neutrons remaining superfluid.
The modeled rise times are consistent with the observed upper limit of 12.6 seconds, and the observed overshoot requires strong crustal friction but weak core friction, supporting a spatially varying \(\mathcal{B}\). These findings highlight the importance of microphysical modeling and demonstrate the potential of future high-cadence timing observations to further constrain the internal dynamics and composition of neutron stars.

The hierarchical nature of galaxy formation in the ΛCDM framework often leads to the formation of multiple supermassive black holes (SMBHs) in the galactic nuclei. The timescale over which galaxies merge, plays a crucial role in shaping the dynamical evolution and the merger dynamics of their central SMBHs, which may result in the formation of a triple (or multiple) SMBH system. While the evolution of binary SMBH is well studied, the long-term dynamics of triple SMBH systems, particularly in non-spherical potentials, still remain less understood. In this work, we investigate the impact of triaxial shape of galaxies in the dynamical evolution of triple SMBHs with initial conditions drawn from the ROMULUS25 cosmological simulation, using different high-resolution gravitodynamical N-body simulations. We explore different orbital configurations and host galaxy shapes, tracking the orbital evolution from a galactic inspiral phase to the formation of hard binaries at sub-parsec separation, and use the observed hardening rates to project the time of coalescence. In all cases, the two most massive SMBHs form a rapidly hardening binary that coalesces within a fraction of a Hubble time, while the third black hole either forms a stable hierarchical triple system with the heavier binary, or remains on a wide orbit.

Typically, pressure inside a neutron star is assumed to be isotropic. However, pressure anisotropy, the difference in pressure between the radial and tangential directions, may arise from strong magnetic fields, viscosity, and elasticity. Neutron-star quasi-normal modes are oscillations with significant damping over time. These modes are classified according to their restoring force. Previous studies showed that pressure modes (p-modes) can become unstable when the neutron star becomes anisotropic. This drove us to study the behaviour of another mode, the w-mode, in anisotropic neutron stars. W-modes are driven mostly by spacetime oscillations with only minor matter perturbations. In this talk, I report the first ever calculation of anisotropic w-modes. Unlike p-modes, we found w-modes to be stable for multiple equation of states in the presence of pressure anisotropy. We have also discovered an equation-of-state-independent relation for w-mode frequency that may potentially help with probing the neutron-star interior without knowing its exact equation of state. This numerical stability finding is further supported by a semi-analytic analysis that ensures stability beyond the scope of the numerical work. We have also found that an approximation scheme that is known to work well for isotropic neutron stars w-modes ceases to be valid when pressure anisotropy is introduced.