"Research objectives and content: we propose to apply recently developed non-perturbative techniques for the simulation of two-flavour QCD with sea quark masses at physical values in order to investigate the convergence properties of Chiral Perturbation Theory (ChiPT) in pi-pi scattering and other light-quark physics quantities. Our approach is based on the Wilson formulation of the fermionic QCD action and an original modification of the Domain Decomposition Hybrid Monte Carlo (DD-HMC) algorithm. Target observables are: pi-pi scattering lenghts, pion mass, pion decay constant, other channels of the light-hadron spectrum. As a result, we expect to find conclusive quantitative information concerning the convergence radius of the chiral expansion and to consequently improve the understanding of the sponaneous breaking of chiral symmetry. Methodology: the main obstacle to simulate light-quarks at physical masses is represented by instability and non-ergodicity phenomena in the HMC, related to the structure of the low-end of the Wilson-Dirac spectrum. We propose to adopt a modification of the algorithm where the low-modes are separated from the bulk of the modes and included in a reweighting factor. Training: the candidate would like to acquire and deepen technical skills and knowledge in the areas of : a) Chiral Perturbation Theory; b) phenomenology of the light hadron interactions; c) Monte Carlo simulations of lattice QCD; d) Parallel computing."