Seminar Details

Home > Seminars > Details

2026-09-22 (11:00) : Ferroelectric Field-Effects with Hafnium Oxide for Neuromorphic Hardware

At Shannon

Duration: 75 minutes

Organized by Institute of Information and Communication Technologies, Electronics and Applied Mathematics

Speaker : Laura Bégon-Lours (ETH Zürich)

Laura Bégon-Lours

Prof. Laura Bégon-Lours is Professor of Neuromorphic Electronics with Oxides of the Dept. of Information Technology and Electrical Engineering at ETH Zürich. She received an Engineer Degree from ESPCI Paris-PSL in 2012, a MSc degree from Ecole Polytechnique in 2012, and a Ph.D. in Physics in 2017 from Université Pierre et Marie Curie. Prof. Bégon-Lours’ research interests include Ferroelectric Thin Films, Hafnium Zirconium Oxide, Clean Room Fabrication, Electrical Characterization, with applications in Artificial Neural Networks Accelerators and Neuromorphic Engineering. She has received several awards, including the MSCA Individual Fellowship (carried at IBM Research Zürich) and the SNFS Starting Grant. Currently, she serves as the chair of the Swiss Chapter of the IEEE Circuits and Systems Society and the IEEE Electron Devices Society.

Abstract : In ferroelectric resistive weights, the strength of the synaptic connection between two neurons is stored in the device conductance. During learning, programming pulses are applied to the synaptic weight, which reconfigures the ferroelectric domains and adjusts the conductance. One strategy to lower the energy cost during the training phase is to lower the duration of the programming pulses. However, the latter cannot be shorter than the self-loading time of the resistive weights, limited by parasitic delays in the circuits. We fabricate ferroelectric resistive weights using bilayers based on hafnia/zirconia superlattices and tungsten oxide. Using this process, CMOS Back-End-Of-Line integration was demonstrated. We determine the maximal device area for which the self-loading time becomes sufficiently short to enable 20 ns programming, which corresponds to a maximum of 3 pJ per pulse. We show that spiking neural network can be deployed on these devices for adaptive electroencephalography decoding. Finaly, ferroelectric capacitors based on the same material also exhibit low-power programming and fast switching speed: full ferroelectric domain reversal is obtained for 5V pulses of only 1 ns.

← Back to Seminars