Description:Collective excitations in quantum many-body systems
Abstract:It is a well known fact that the quantum many-body problem can not be solved exactly and one has to resort to approximate methods to study their properties. Moreover, modern studies of ground state and excited states of many body systems require more and more powerful computer resources. The GRID infrastructure is thus a particularly well suited tool for their implementation.
This work is directed to the development of an extension of the Random Phase Approximation (RPA) aimed to give a better description of ground state correlations. RPA is a microscopic approach largely used in the study of collective excitations in many-body systems and is usually carried out in order to improve the standard mean field description of a physical system, by introducing some correlations which are neglected at mean field level. However, RPA presents some limits and many efforts are still done by the scientific community in order to overcome them by improving the RPA description of ground state correlations. In such a line of investigation, we have developed an improved RPA approach in order to take in account correlations in a self-consistent way. From a numerical point of view, the improved RPA equations have the same form of the RPA ones but are non-linear. So we implement a code in which we solve them via an iterative procedure. We have applied this approach in the analysis of the electronic properties of Metal Clusters. In particular we have analysed the dipole plasmon modes and we have found a good agreement with experimental results, better than standard RPA does. By using the PI2S2 resources, it has been possible to carry out this analysis for several Sodium Metal Clusters, both neutral and ionized. We remark that, at each step of the iterative procedure, one has to solve the \’’improved RPA\’’ equations for several multipolarities, both for spin =0 and 1 channels and this makes the calculations more time consuming that the ones of the Standard RPA. The possibility of using many CPUs at the same time (trivial paralleliza