Constraints on dark energy and the future lifetime of the universe.

摘要

Since the discovery of the acceleration of the expansion of the universe, the nature of the energy density driving the acceleration has remained a mystery. Dark energy could arise from a slowly rolling scalar field, similar to inflation in the early universe but on a much smaller energy scale. We use present and future observations to give bounds on the equation of state and on the model parameters directly, and calculate the minimal future lifetime of the universe still consistent with observations. In the simplest model of this type, the linear potential V(&phis;) = V0(1 + alpha&phis;), current observations allow a collapse to occur in as little as 24 billion years. Future observations in the next decade may push this bound to 40 billion years. We further study the generic shape of the constraints on model parameters obtained with the Markov Chain Monte Carlo (MCMC) method and explain features, which seem to imply nontrivial dynamics, as an accumulation of probability due to marginalization over other parameters. We also show that for a dark energy model from gauged N=8 supergravity recollapse may occur in as early as 8 billion years, which is less than the total lifetime of the sun. In its very simplest form, dark energy could be vacuum energy density and the universe would continue to expand forever. Using the Fisher matrix method, we assesses how well future observations could exclude this possibility.; Reheating after inflation in the early universe has been studied in recent years in all but the first of the three main types of inflationary models, new inflation, chaotic inflation, and hybrid inflation. We have completed this study here by investigating reheating after new inflation, using analytical methods and lattice simulations. Surprisingly, both tachyonic preheating and parametric resonance are important and contribute comparably to the decay of the zero mode of the inflaton field and subsequent particle production.; The final chapter of this thesis is devoted to a technical presentation of a computational technique to simulate eternal inflation using linear lists.