Cosmology and Gravitation

Realizations of inflation in string theory hold the promise of connecting the theory to observational tests, and at the same time providing new insights for field theory models of inflation. I will report on progress towards realizing inflation on D-branes in type IIB string theory. Moduli stabilization effects generically lead to an eta problem in this scenario, and to analyze the model it is necessary to compute a particular correction to the nonperturbative effects arising on wrapped D-branes. I will explain this calculation, then present results for the full inflaton potential that establish the existence of successful models, albeit with qualitatively new predictions.

I argue that all necessary ingredients for successful inflation are present in the minimal supersymmetric standard model (MSSM). The potential for the supersymmetric flat directions (which can be viewed as moduli near points of enhanced symmetry) generically has metastable minima at large field values. An adjustment of soft supersymemtry breaking parameters results in a point where the first two derivatives of the potential vanish. One can then have more than 10^3 e-foldings of slow roll inflation with a Hubble rate ~ O(1-10) GeV in the vicinity of this point. The model has a robust prediction for the scalar spectral index: n_s \leq 0.93, which agrees with the current WMAP and LSS data within 2\sigma. It can be implemented purely within MSSM, or MSSM plus Dirac neutrinos. It has the unique feature of having a concrete and real connection to physics that can be observed in earthbound laboratories.

I will demonstrate how one can realize Cascade inflation in M-theory. Cascade inflation is a realization of assisted inflation which is driven by non-perturbative interactions of N M5-branes. Its power spectrum possesses three distinctive signatures: a decisive power suppression at small scales, oscillations around the scales that cross the horizon when the inflaton potential jumps and stepwise decrease in the scalar spectral index. All three properties result from features in the inflaton potential. The features in the inflaton potential are generated whenever two M5-branes collide with the boundaries. The derived small-scale power suppression serves as a possible explanation for the dearth of observed dwarf galaxies in the Milky Way halo. The oscillations, furthermore, allow to directly probe M-theory by measurements of the spectral index and to distinguish cascade inflation observationally from other string inflation models.

In this talk I will discuss some aspects of graviton production by moving branes. After a brief introduction to braneworld cosmology I will focus on braneworlds in a five-dimensional bulk, where cosmological expansion is mimicked by motion through AdS_5. The moving brane acts naturally as a time-dependent boundary for the five-dimensional graviton (five-dimensional tensor perturbations) leading to graviton production out of quantum vacuum fluctuations. This effect is related to the so-called dynamical Casimir effect, i.e. the generation of real photons out of vacuum fluctuations of the quantized electromagnetic field in dynamical cavities. By applying the formalism used to study the dynamical Casimir effect I will show explicitly that the five-dimensional graviton reduces to the four-dimensional one in the late time approximation of such braneworlds. In the last part of the talk I will study a (toy) model where two branes approach each other in a radiation dominated phase, bounce off and move apart from each other afterwards. Thereby generation of massive gravitons takes place caused by the coupling of the Kaluza-Klein modes to the gravitational zero mode which exhibits a blue spectrum. At the end I will discuss possible applications of the formalism to more interesting scenarios (braneworld inflation etc).

The best studied class of dark matter candidates in Supersymmetric theories is the WIMP, Weakly Interacting Massive Particles, which makes cold dark matter. There is a well-motivated alternative to the WIMP -- dark matter populated by decays of WIMPs. This dark matter from decays is closer in spirit to warm dark matter. They can be distinguished from cold dark matter by observations of structure on scales smaller than about a megaparsec, where cold dark matter models seem to face difficulty. Big Bang Nucleosynthesis predictions are also modified in interesting ways.

Experiments have ruled out unit-strength scalar-mediated fifth forces on scales ranging from 0.1 mm to 10,000 AU. However, allowing the scalar to have a quartic self-interaction weakens these constraints considerably. This weakening is due to the "chameleon mechanism", which gives the scalar field an effective mass that depends on the local matter density. I will describe the chameleon mechanism and discuss experimental constraints on self-interacting scalar fields. In particular, I will compare the chameleon-mediated self interaction to constraints from the Eot-Wash experiment, at the University of Washington, which comes closest to detecting such a scalar field today. It will be shown that a quartic self interaction of unit strength is just out of reach of the current Eot-Wash experiment, but will be readily visible to their next-generation instrument.

It has recently been proposed by Nayeri, Brandenberger and Vafa, that the thermodynamics of strings in the early universe can provide us with a causal mechanism to generate a scale invariant spectrum of primordial density fluctuations, without requiring an intervening epoch of inflation. We will review this mechanism, and report on more recent work which has uncovered several observational consequences of the NBV mechanism, some of which in principle, will be distinguishable from the generic predictions of inflation.

We propose a new brane world scenario. In our model, the Universe starts as a small bulk filled with a dense gas of branes. The bulk is bounded by two orbifold fixed planes. An initial stage of isotropic expansion ends once a weak potential between the orbifold fixed planes begins to dominate, leading to contraction of the extra spatial dimensions. Depending on the form of the potential, one may obtain either a non-inflationary scenario which solves the entropy and horizon problem, or an improved brane-antibrane inflation model.

A cosmological model based on an inhomogeneous D3-brane moving in an AdS_5 X S_5 bulk is introduced. Although there is no special points in the bulk, the brane Universe has a center and is isotropic around it. The model has an accelerating expansion and its effective cosmological constant is inversely proportional to the distance from the center, giving a possible geometrical origin for the smallness of a present-day cosmological constant. Besides, if our model is considered as an alternative of early time acceleration, it is shown that the early stage accelerating phase ends in a dust dominated FRW homogeneous Universe. Mirage-driven acceleration thus provides a dark matter component for the brane Universe final state. We finally show that the model fulfills the current constraints on inhomogeneities.

I discuss the backreaction of inhomogeneities on the expansion of the universe. The average behaviour of an inhomogeneous spacetime is not given by the Friedmann-Robertseon-Walker equations. The new terms in the exact equations hold the possibility of explaining the observed acceleration without a cosmological constant or new physics. In particular, the coincidence problem may be solved by a connection with structure formation.