We present a model for the metal–insulator transition in 2D, observed in the recent years. Our starting point consists of two ingredients only, which are ubiquitous in the experiments: Coulomb interactions and weak disorder spin scattering (coming from the interfaces of the heterostructures in question). In a diagramatic approach, we predict the existence of a characteristic temperature T 0 = T 0 ( n , ω H ) , n being the density of carriers, and ω H the Zeeman energy, below which these systems become metallic, due to the onset of strong spin–density correlations. This is in very good agreement with experiments, and corroborates the fact that varying n and ω H are equivalent ways into/out of the metallic regime. The conductivity, calculated as a function of temperature and ω H in the metallic state, compares favorably to experiment. Moreover, we give an explicit expression for the conventional weak disorder contributions to the conductivity in the frame of our model. We comment on the nature of the transition, we calculate the specific heat of the system and we discuss the fate of the metallic state in the limit of zero temperature.
Read full abstract