Submergence and re-diffusion of the neutron star magnetic field after the supernova

We consider the effect of post core-collapse accretion on the magnetic field of the new-born neutron star. If this accretion is hypercritical then the ram pressure overwhelms the magnetic field pressure and we show that in this case the accretion induced flow time scale in the upper layers of the neutron star is shorter, by orders of magnitude, than the magnetic field ohmic diffusion time scale. This means that the magnetic field is frozen in the matter and any initial magnetic field of the neutron star is rapidly submerged beneath the accreted matter. If the accreting matter is weakly, pr non, magnetized, this implies that neutron stars produced by supernovae in which hypercritical accretion occurred are born with weak, or even vanishing, surface magnetic field. Later diffusion of the magnetic field back to the surface could produce a delayed switch-on of a pulsar (Muslimov & Page 1995): We model this re-diffusion in detail for a wide range of submergence depths and discuss the consequences for pulsar observables, such as the period P and its time derivative P. As a result of the field submergence, the spin-down age tau(sd) can be much larger than the real age of the pulsar. Moreover, we show that if the field submergence is deep enough the magnetic field will be hidden for many millions of years. This mechanism of field submergence may explain the lack of evidence for the presence of a pulsar in all recent supernovae and may also contribute to the discrepancy between the estimated pulsar birth rate and type Ib+II supernova rate. In particular, we predict that a pulsar will never turn-on in the remnant of SN 1987A.