Astrophysical and cosmological doomsdays

In this dissertation we study two well known gravitational scenarios in which singularities may appear; the final state of gravitational collapse and the late time evolution of the universe. In the first scenario, we study a spherically symmetric space-time whose matter content includes a tachyon scalar field and a barotropic fluid. By employing a dynamical system analysis, we find classical solutions corresponding to a naked singularity or a black hole formation. We then investigate, in a semiclassical manner, loop quantum gravity induced effects on the fate of the classical singularities. By employing an inverse triad correction, we identify a subset which corresponds to an outward flux of energy, thus avoiding either a naked singularity or a black hole formation. Within a holonomy correction, we obtain the semiclassical counterpart of our (classical) general relativistic collapse in which, classical singularity is resolved and replaced by a bounce. In addition, we find a threshold scale for non-singular black hole formation. In the second scenario, we employ several models of dark energy to investigate the status of the late time cosmological singularities. In the first one we consider a DGP brane-world model with a GB term being provided for the bulk action; a phantom matter is present on the brane as dark energy component. It is shown that a combination of IR and UV modifications to general relativity replaces a big rip by a sudden singularity at late times. The second model we cosnider is the generalized running vacuum energy (GRVE) model. The Friedmann equation of the GRVE model looks much similar to the one of a holographic Ricci dark energy (HRDE) model. Despite the analogy between these two models, it turns out that one of them, a GRVE, is singularity-free in the future while the other, the HRDE, is not, which can hit, for example, a big rip singularity