Until now the forecast of the Semmiconductors Industry Association (SIA) concerning the dimension shrinking and the performance improvement of the electrical devices, reported in the International Technology Roadmap for Semiconductors (ITRS), matched very precisely the development of semiconductor process technology. But today the traditional scaling is indeed approaching the fundamental limits of the materials consituting the building blocks of the CMOS process. A big and unresolved challenge in the traditional process shrinking approach is the gate insulator. To be able to follow the dimension shrinking according to the ITRS, the SiO2 film thickness should become below 1nm within the next three years. This thickness corresponds to few atomic layers, which means that the direct tunnel leakage current through the insulator will increase. The high leakage current and the inadequate reliability for a SiO2 layer of less than 1.5nm thickness require a replacement for SiO2. To obtain high gate capacitance and inhibit tunneling, relative thick insulator of high dielectric constant (high-k) are needed to replace silicon dioxide (SiO2) as gate oxide. Therefore new materials have to be introduced into the basic CMOS structure to replace the existing ones to further extend device scaling and the reduction of the produciont costs. The present research thesis focuses on the proposition and investigation of three alternative gate oxide systems: aluminium-, praseodymium- and lanthanum oxide (Al2O3, Pr2O3 and La2O3 respectively). For each one of these systems, the growth process by Molecular Beam Deposition (MBD) has been optimised and electrical and physical characterisation has been performed to gain a better understanding of important factors associated with alternative gate dielectrics form both a theoretical and experimental point of view. Moreover, the optimisation of the interface between gate dielectric and the silicon substrate is taken into account during the development of the deposition processes. The first part of the thesis concerns the aluminium oxide. Aluminium oxide (Al2O3) is one of the first systems which have been studied to replace silicon dioxide as gate dielectric because of its large barrier height, dielectric constant twice that of SiO2, high stability and robustness. The basic properties of Al2O3 films grown on silicon substrate are well understood and for this reason alumina can be used as reference to investigate on new materials for alternative gate oxide. Beyond the aluminium oxide, lanthanide oxides have been considered as long term solution to the high-k question. In particular preseodymium oxide (Pr2O3) and lanthanum oxide (La2O3) have attracted the attention because of their high dielectric constant (20-30) and thermal stability on silicon substrate until 1000K. The properties of thin lanthanide oxide films as dielectric system for microelectronic applications are not yet completely known ind intensive research is running to find out if this dielectric will cover all the requirements needed for the new gate oxide material. In particular the major drawback of lanthanide oxide is given by its high sensibility to humidity, which leads to degradation of the dielectric film. This thesis will try to give an answer to the open questions on the investigated materials and will show the direction for future investigations.    «

Until now the forecast of the Semmiconductors Industry Association (SIA) concerning the dimension shrinking and the performance improvement of the electrical devices, reported in the International Technology Roadmap for Semiconductors (ITRS), matched very precisely the development of semiconductor process technology. But today the traditional scaling is indeed approaching the fundamental limits of the materials consituting the building blocks of the CMOS process. A big and unresolved challenge i...    »