Revise thermal winds of remnant neutron stars in gamma-ray bursts

It seems that the wealth of information revealed by the multi-messenger observations of the binary neutron star (NS) merger event, GW170817/GRB 170817A/kilonova AT2017gfo, places irreconcilable constraints to models of the prompt emission of this gamma-ray burst (GRB). The observed time delay between the merger of the two NSs and the trigger of the GRB and the thermal tail of the prompt emission can hardly be reproduced by these models simultaneously. We argue that the merger remnant should be an NS (last for, at least, a large fraction of 1s), and that the difficulty can be alleviated by the delayed formation of the accretion disk due to the absorption of high-energy neutrinos emitted by the NS and the delayed emergence of an effective viscous in the disk. Further, we extend the consideration of the effect of the energy deposition of neutrinos emitted from the NS. If the NS is the central object of a GRB with a distance and duration similar to that of GRB 170817A, thermal emission of the thermal bubble inflated by the NS after the termination of accretion may be detectable. If our scenario is verified, it would be of interest to investigate the cooling of nascent NSs.