Abstract:
An attempt is made of giving a self-contained introduction to holomorphic ideas in general relativity, following work over the last thirty years by several authors. The main topics are complex manifolds, spinor and twistor methods, heaven spaces.

Abstract:
A recent analysis of real general relativity based on multisymplectic techniques has shown that boundary terms may occur in the constraint equations, unless some boundary conditions are imposed. This paper studies the corresponding form of such boundary terms in complex general relativity, where space-time is a four-complex-dimensional complex-Riemannian manifold. A complex Ricci-flat space-time is recovered providing some boundary conditions are imposed on two-complex-dimensional surfaces. One then finds that the holomorphic multimomenta should vanish on an arbitrary three-complex-dimensional surface, to avoid having restrictions at this surface on the spinor fields which express the invariance of the theory under holomorphic coordinate transformations. The Hamiltonian constraint of real general relativity is then replaced by a geometric structure linear in the holomorphic multimomenta, and a link with twistor theory is found. Moreover, a deep relation emerges between complex space-times which are not anti-self-dual and two-complex-dimensional surfaces which are not totally null.

Abstract:
In complex general relativity, Lorentzian space-time is replaced by a four-complex-dimensional complex-Riemannian manifold, with holomorphic connection and holomorphic curvature tensor. A multisymplectic analysis shows that the Hamiltonian constraint is replaced by a geometric structure linear in the holomorphic multimomenta, providing some boundary conditions are imposed on two-complex-dimensional surfaces. On studying such boundary conditions, a link with the Penrose twistor programme is found. Moreover, in the case of real Riemannian four-manifolds, the local theory of primary and secondary potentials for gravitino fields, recently proposed by Penrose, has been applied to Ricci-flat backgrounds with boundary. The geometric interpretation of the differential equations obeyed by such secondary potentials is related to the analysis of integrability conditions in the theory of massless fields, and might lead to a better understanding of twistor geometry. Thus, new tools are available in complex general relativity and in classical field theory in real Riemannian backgrounds.

Abstract:
In this paper it is implemented how to make compatible the boundary conditions and the gauge fixing conditions for complex general relativity written in terms of Ashtekar variables using the Henneaux-Teitelboim-Vergara approach. Moreover, it is found that at first order in the gauge parameters, the Hamiltonian action is (on shell) fully gauge-invariant under the gauge symmetry generated by the first class constraints in the case when spacetime $\cal M$ has the topology ${\cal M}= R \times \Sigma$ and $\Sigma$ has no boundary. Thus, the statement that the constraints linear in the momenta do not contribute to the boundary terms is right, but only in the case when $\Sigma$ has no boundary.

Abstract:
The Plebanski formulation of complex general relativity is given in terms of variables valued in the complexification of the $so(3)$ Lie algebra. Therefore, it is genuinely a gauge theory that is also diffeomorphism-invariant. For this reason, the way that the Levi-Civita connection emerges from this formulation is not direct because both the internal (gauge) and the spacetime connections are geometrical structures a priori not related, there is not a natural link between them. Any possible relationship must be put in by hand or must come from extra hypotheses. In this paper, we analyze the correct relationship between these connections and show how they are related.

Abstract:
This thesis is based on three different projects, all of them are directly linked to the classical general theory of relativity, but they might have consequences for quantum gravity as well. The first chapter deals with pseudo-Finsler geometric extensions of the classical theory, these being ways of naturally representing high-energy Lorentz symmetry violations. The second chapter deals with the problem of highly damped quasi-normal modes related to different types of black hole spacetimes. Besides the astrophysical meaning of the quasi-normal modes, there are conjectures about the link between the highly damped modes and black hole thermodynamics. The third chapter is related to the topic of multiplication of tensorial distributions.

Abstract:
In this paper, using E. Carten's exterior calculus, we give the spinor form of the structure equations, which leads naturally to the Newman-Penrose equations. Further, starting from the spinor space and the sl(2C) algebra, we construct the general complex-vector formalism of general relativity. We find that both the Cahen-Debever-Defrise complex-vector formalism and the Brans one are its special cases. Thus, the spinor formalism and the complex-vector formalism of general relativity are unified on the basis of the unimodular group SL(2C) and its Lie algebra.

Abstract:
Using the Moller energy--momentum definition in general relativity(GR) we calculate the total energy--momentum distribution associated with (n+2)-dimensional homogeneous and isotropic model of the universe. It is found that total energy of Moller is vanishing in (n+2) dimensions everywhere but n-momentum components of Moller in (n+2) dimensions are different from zero. Also, we evaluate the static Einstein Universe, FRW universe and de Sitter universe in four dimensions by using (n+2)-type metric, then calculate the Moller energy--momentum distribution of these spacetimes. However, our results are consistent with the results of Banerjee and Sen, Xulu, Radinschi,Vargas, Cooperstock-Israelit, Aygun et al., Rosen, and Johri et al. in four dimensions.

Abstract:
This document contains a description of physics entirely based on a geometric presentation: all of the theory is described giving only a pseudo-riemannian manifold (M, g) of dimension n > 5 for which the g tensor is, in studied domains, almost everywhere of signature (-, -, +, ..., +). No object is added to this space-time, no general principle is supposed. The properties we impose to some domains of (M, g) are only simple geometric constraints, essentially based on the concept of "curvature". These geometric properties allow to define, depending on considered cases, some objects (frequently depicted by tensors) that are similar to the classical physics ones, they are however built here only from the g tensor. The links between these objects, coming from their natural definitions, give, applying standard theorems from the pseudo-riemannian geometry, all equations governing physical phenomena usually described by classical theories, including general relativity and quantum physics. The purely geometric approach introduced hear on quantum phenomena is profoundly different from the standard one. Neither Lagrangian nor Hamiltonian are used. This document ends with a quick presentation of our approach of complex quantum phenomena usually studied by quantum field theory.