Classical limit of the Nelson model

Since the development of quantum mechanics it has been natural to analyze the connection between classical and quantum mechanical descriptions of physical systems. In particular one should expect that in some sense when quantum mechanical effects becomes negligible the system will behave like it is dictated by classical mechanics. One famous relation between classical and quantum theory is due to Ehrenfest. This result was later developed and put on firm mathematical foundations by Hepp. He proved that matrix elements of bounded functions of quantum observables between suitable coherents states (that depend on Planck's constant h) converge to classical values evolving according to the expected classical equations when h goes to zero. His results were later generalized by Ginibre and Velo to bosonic systems with infinite degrees of freedom and scattering theory. In this thesis we study the classical limit of Nelson model, that describes non relativistic particles, whose evolution is dictated by Schrödinger equation, interacting with a scalar relativistic field, whose evolution is dictated by Klein-Gordon equation, by means of a Yukawa-type potential. The classical limit is a mean field and weak coupling limit. We proved that the transition amplitude of a creation or annihilation operator, between suitable coherent states, converges in the classical limit to the solution of the system of differential equations that describes the classical evolution of the theory. The quantum evolution operator converges to the evolution operator of fluctuations around the classical solution. Transition amplitudes of normal ordered products of creation and annihilation operators between coherent states converge to suitable products of the classical solutions. Transition amplitudes of normal ordered products of creation and annihilation operators between fixed particle states converge to an average of products of classical solutions, corresponding to different initial conditions.

Electrothermal characterization and nonlinear modelling of GaN transistors for telecommunication circuit design

This thesis starts showing the main characteristics and application fields of the AlGaN/GaN HEMT technology, focusing on reliability aspects essentially due to the presence of low frequency dispersive phenomena which limit in several ways the microwave performance of this kind of devices. Based on an equivalent voltage approach, a new low frequency device model is presented where the dynamic nonlinearity of the trapping effect is taken into account for the first time allowing considerable improvements in the prediction of very important quantities for the design of power amplifier such as power added efficiency, dissipated power and internal device temperature. An innovative and low-cost measurement setup for the characterization of the device under low-frequency large-amplitude sinusoidal excitation is also presented. This setup allows the identification of the new low frequency model through suitable procedures explained in detail. In this thesis a new non-invasive empirical method for compact electrothermal modeling and thermal resistance extraction is also described. The new contribution of the proposed approach concerns the non linear dependence of the channel temperature on the dissipated power. This is very important for GaN devices since they are capable of operating at relatively high temperatures with high power densities and the dependence of the thermal resistance on the temperature is quite relevant. Finally a novel method for the device thermal simulation is investigated: based on the analytical solution of the tree-dimensional heat equation, a Visual Basic program has been developed to estimate, in real time, the temperature distribution on the hottest surface of planar multilayer structures. The developed solver is particularly useful for peak temperature estimation at the design stage when critical decisions about circuit design and packaging have to be made. It facilitates the layout optimization and reliability improvement, allowing the correct choice of the device geometry and configuration to achieve the best possible thermal performance.