The FUNCHIP project aims to determine the physical limits of parallel plate electrostatic actuators used in MEMS RF switches, and optimize the thermal behavior of these components. Applications targeted in this project include active antennas for Radars, in which RF - MEMS relays would protect the RF front heads efficiently, while having reduced losses. Technological roadbloacks concern power handling , which must be greater than 20 Watts between 6 and 20 GHz , with over 1 Billion cycles reliabilities . With the maturation of research in this area, the keys for achieving such performance are emerging. The electrostatic actuator which is used for opening and closing the relays must generate large forces. This allows using very stiff, fast structures, and insensitive to adhesion. Furthermore, the resistance between the two electrodes when the relay is closed, is lower when the pressure is important. For a given power level , the temperature elevation at the contact point is related to the resistance of this point (and therefore to the force generated by the actuator ) , but also to the ability of the substrate to remove the heat generated . The FUNCHIP project addresses these two aspects , the actuator, and thermal managment ( packaging and the switching time will not be addressed in this project). For a given voltage , the electrostatic actuators parallel plate MEMS see their strength increase progressively as the gap separation between the plates decreases. Thus, small separation gaps permit reaching large forces. This increase is limited byelectrical breakdown phenomena that not well understood today because the physics of submicron breakdown ( Paschen effect , for example) has been little studied . Beyond the effects of the ambient gas , the nature of the metals used for the electrodes, their roughness , and layout affects their behavior. The FUNCHIP project will investigate systematically the properties of metals commonly used in microelectronics and test the breakdown strength of several simple test vehicles. Leakage current measurements will also look more finely into the physical phenomena involved in order to determine the nature of these currents. We will be able to determine gaps heights that ensure a reliable and optimal operation of these actuators. To optimize the thermal behavior of the relay , substrate transfer techniques will be used. Using thin substrates will directly improve thermal performance of micro- switches. However, these thin substrates will reduce the cross section of access lines to micro- switches, and therefore increase losses, and the mismatch between the switch structure and input lines. Trhough electromagnetic and thermal simulations simulations, we will find an optimal compromise to achieve a component with good thermal characteristics and low losses. The last part of the project will be devoted to design and implementation of a micro switch following the previous results. Specifically, we will build a micro switch with a powerful electrostatic actuator, with a gap as small as possible, on an optimized substrate. The micro -switch will be tested at high microwave powers to targeted levels from thermal simulations. With available imaging techniques to LAAS , we can validate our simulations and have effective design rules for power applications adapted to defense applications .