Is the 3-D magnetic null point with a convective electric field an efficient particle accelerator?

Aims. We study the particle acceleration at a magnetic null point in the solar corona, considering self-consistent magnetic fields, plasma flows and the corresponding convective electric fields. Methods. We calculate the electromagnetic fields by 3-D magnetohydrodynamic (MHD) simulations and expose charged particles to these fields within a full-orbit relativistic test-particle approach. In the 3-D MHD simulation part, the initial magnetic field configuration is set to be a potential field obtained by extrapolation from an analytic quadrupolar photospheric magnetic field with a typically observed magnitude. The configuration is chosen so that the resulting coronal magnetic field contains a null. Driven by photospheric plasma motion, the MHD simulation reveals the coronal plasma motion and the self-consistent electric and magnetic fields. In a subsequent test particle experiment the particle energies and orbits (determined by the forces exerted by the convective electric field and the magnetic field around the null) are calculated in time. Results. Test particle calculations show that protons can be accelerated up to 30 keV near the null if the local plasma flow velocity is of the order of 1000 km s-1 (in solar active regions). The final parallel velocity is much higher than the perpendicular velocity so that accelerated particles escape from the null along the magnetic field lines. Stronger convection electric field during big flare explosions can accelerate protons up to 2 MeV and electrons to 3 keV. Higher initial velocities can help most protons to be strongly accelerated, but a few protons also run the risk to be decelerated. Conclusions. Through its convective electric field and due to magnetic nonuniform drifts and de-magnetization process, the 3-D null can act as an effective accelerator for protons but not for electrons. Protons are more easily de-magnetized and accelerated than electrons because of their larger Larmor radii. Notice that macroscopic MHD simulations are blind to microscopic magnetic structures where more non-adiabatic processes might be taking place. In the real solar corona, we expect that particles could have a higher probability to experience a de-magnetization process and get accelerated. To trigger a significant acceleration of electrons and even higher energetic protons, however, the existence of a resistive electric field mainly parallel to the magnetic field is required. A physically reasonable resistivity model included in resistive MHD simulations is direly needed for the further investigations of electron acceleration by parallel electric fields