We revisit in this work the problem of the maximum masses of magnetized White Dwarfs (WD). The impact of a strong magnetic field onto the structure equations is addressed. The pressures become anisotropic due to the presence of the magnetic field and
split into a parallel and perpendicular components. We first construct stable solutions of TOV equations for the parallel pressures, and found that physical solutions vanish for the perpendicular pressure when $B gtrsim 10^{13}$ G. This fact establishes an upper bound for a magnetic field and the stability of the configurations in the (quasi) spherical approximation. Our findings also indicate that it is not possible to obtain stable magnetized WD with super Chandrasekhar masses because the values of the magnetic field needed for them are higher than this bound. To proceed into the anisotropic regime, we derived structure equations appropriated for a cylindrical metric with anisotropic pressures. From the solutions of the structure equations in cylindrical symmetry we have confirmed the same bound for $B sim 10^{13} $ G, since beyond this value no physical solutions are possible. Our tentative conclusion is that massive WD, with masses well beyond the Chandrasekhar limit do not constitute stable solutions and should not exist.
We investigate the effects of the anomalous magnetic moment (AMM) in the EoS of a fermion system in the presence of a magnetic field. In the region of strong magnetic fields ($B>m^2$) the AMM is found from the one-loop fermion self-energy. In contras
t to the weak-field AMM found by Schwinger, in the strong magnetic field case, the AMM depends on the Landau level (LL) and decreases with it. The effects of the AMM in the EoS at intermediate-to-large fields can be found introducing the one-loop, LL-dependent AMM in the effective Lagrangian that is then used to find the thermodynamic potential of the system. We compare the plots of the parallel and perpendicular pressures versus the magnetic field in the strong field region considering the LL-dependent AMM, the Schwinger AMM, and no AMM at all. The results clearly show a separation between the physical magnitudes found using the Schwinger AMM and the LL-dependent AMM. This is an indication of the inconsistency of considering the Schwinger AMM beyond the weak field region $B< m^2$ where it was originally found. The curves for the EoS, pressures and magnetization at different fields give rise to the well-known de Haas van Alphen oscillations, associated to the change in the number of LL contributing at different fields.
