The development in the early 2000s of non-volatile resistive memories, known as memristors, has enabled the design of energy-efficient devices. The recent industrialisation of microsystems using this type of memory is proof of the interest in this type of device. In addition, the development of wireless telecommunications systems is leading to the use of frequency ranges that have been little exploited until now. In order to maintain the desired performance in these frequency ranges, while continuing to increase the density of electronic microsystems, it is necessary to develop compact devices that operate at high frequencies.
For this last point, 2D materials are ideal candidates. The physical properties of these materials remain unchanged even for a single atomic layer. Switching voltages are lower and cut-off frequencies are higher. In addition, various research projects have demonstrated the memristive behaviour of MoS2, a 2D semiconductor material. However, the physical phenomena occurring within memristors made from MoS2 are not yet fully understood, and the performance achieved by these devices is currently insufficient for them to be integrated into commercial devices. This is because the architecture of these devices requires the upper electrode to completely cover the MoS2, which prevents physico-chemical characterisation of the 2D material.
During this thesis work, a process was developed to exfoliate the upper electrode and thus allow access to the MoS2 for surface characterisation tools. The MoS2 could thus be characterised in the different electrical states of the memristor. The evolution of electrical parameters such as endurance, ION/IOFF ratios, switching times and cut-off frequency of MoS2-based memristors was studied as a function of the thickness of the MoS2, its growth method and the nature of the electrodes in contact with the MoS2. This study, combined with results already obtained in the literature, has enabled conclusions to be drawn about the switching mechanisms of memristors and key parameters for device optimisation to be determined

​ ​ ​​