This thesis is a theoretical work on the solid state using computational physics. The basic tool underlying most of the calculations is Density Functional Theory, which describes the system in terms of the electronic density and maps the interacting system onto an equivalent non-interacting set of particles in an effective potential. These simplifications allow us to perform accurate and fast simulations of the ground state physical properties of crystalline solids and isolated systems with a relatively large number of atoms. Such simulations are known as ab initio or first principles simulations because no free parameters enter into them, everything is derived from the postulates of quantum mechanics. However, in practice some parts of the simulation have to be approximated and other parts have to be simplified in order to speed it up and make it computable. This is one of the reasons computational physics is not simply reduced to ‘enter some parameters in the computer and wait for the result’ since prior to the main study one has to carefully determine the most efficient set of ‘simplifications’. Additionally, one has to understand very well the different concepts underlying the basic theory and approximations and know a series of numerical algorithms and computer programs to be able to modify and improve the numerical code. Furthermore, the results have to be carefully analyzed and interpreted and, most importantly, they have to be compared with the experiments and one has to explain why they do or do not match. As I will show, these conditions were completely followed in all the calculations.