Course Announcement,
Ferroelectric-based devices for low-power field-effect transistors and neuromorphic computing, The course consists of 8 hours , and lectures will be on:
September 2nd , 09:00 – 13:00 (Sala Gialla – DPIA – UniUD)
September 3rd , 09:00 – 13:00 (Sala Gialla – DPIA – UniUD)
Daniel Lizzit
Researcher at the Polytechnic Department of Engineering and Architecture (DPIA) – University of Udine
The discovery of ferroelectricity in hafnium oxides in the last decade opened new perspectives for ferroelectric-based transistors. Their applications spread from low-power devices that exploit the negative capacitance effect, to neuromorphic computing thanks to the nonvolatile multilevel memory effect achievable with ferroelectric oxides. It will be given a description of the physical framework to model the behaviour of the ferroelectric oxides. In particular, it will be shown how to get access to the unusual negative capacitance regime in which a destabilization of the ferroelectric polarization induces a negative ferroelectric capacitance, resulting in a differential voltage amplification and thus in a reduced subthreshold swing of the transistor. Ferroelectric thin films oxides are also actively investigated in recent years as consitutive elements of neural networks, and basically as synapses. There will be described the different device solutions based on two or three terminal devices (FTJ and FeFETs) to implement synapses with varying synaptic weights and discussed the different writing and reading conditions.