Abstract

Context. This paper addresses the evolution of an axially symmetric magnetic field in the core of a neutron star. The matter in the core is modeled as a system of two fluids, namely neutrons and charged particles, with slightly di_erent velocity fields controlled by their mutual collisional friction. This problem was addressed in our previous work through the "fictitious friction"approach. Aims. We study the validity of our previous work and improve on it by comparing the fictitious friction approach to alternatives, making approximations that allow it to be applied to arbitrary magnetic field strengths and using realistic equations of state. Methods. We assumed the neutron star crust to be perfectly resistive so that its magnetic field reacts instantaneously to changes in the core, in which we neglect the e_ects of Cooper pairing. We explored di_erent approaches to solve the equations to obtain the velocities and chemical potential perturbations induced by a given fixed magnetic field configuration in the core. We also present a new version of our code to perform time-evolving simulations and discuss the results obtained with it. Results. Our calculations without fictitious friction further confirm that bulk velocity is generally much greater than ambipolar velocity, leading to faster evolution. These findings align with those with fictitious friction, validating this approach. We also find that, in the long term, the star evolves toward a barotropic "Grad-Shafranov equilibrium,"where the magnetic force is fully balanced by charged particle fluid forces. Qualitatively, the evolution and the final equilibrium are independent of the magnetic field strength B and the equation of state considered. The timescale to reach this equilibrium is proportional to B-2 and becomes shorter for equations of state with a smaller gradient of the ratio between the densities of protons and neutrons. © The Authors 2025.