Emulation of synaptic functions in organic ferroelectric tunnel junctions for neuromorphic computing

No Thumbnail Available
Journal Title
Journal ISSN
Volume Title
Sähkötekniikan korkeakoulu | Master's thesis
Date
2019-06-17
Department
Major/Subject
Advanced Materials and Photonics
Mcode
ELEC3035
Degree programme
Master’s Programme in Electronics and Nanotechnology (TS2013)
Language
en
Pages
40
Series
Abstract
Neuromorphic computing is a computing architecture that mimics biological neural systems. Successful implementation of this architecture requires artificial neurons and synapses that reliably replicate selected functions of their biological counterparts. Ferroelectric tunnel junctions (FTJs) can be used as highly tunable and energy efficient synapses. In this work, organic FTJs comprising a P(VDF-TrFE) organic ferroelectric layer sandwiched between Au and Nb-doped SrTiO3 (NSTO) electrodes were fabricated by spin coating of the ferroelectric polymer onto the semiconducting NSTO substrate and evaporation of the Au electrodes. Three samples were prepared on substrates with varying Nb-doping concentrations of 0.25wt%, 0.5wt% and 0.7wt%, and their resistive switching and synaptic properties were characterized by electrical transport measurements at room temperature. Comparisons of the resistive switching data of each sample reveal that the ON-OFF resistance ratio increases with increasing doping concentration due to a modulation of the Schottky barrier at the ferroelectric-semiconductor interface. Furthermore, synaptic potentiation and depression functions are emulated successfully. The experiments also show that the time constants of relaxation after potentiation or depression depend on the Nb doping concentration. This effect is rationalized by improved screening of the depolarization field if the Nb doping concentration is high. The results of this study open a path toward the use of organic FTJs in neuromorphic computing devices.
Description
Supervisor
van Dijken, Sebastiaan
Thesis advisor
Majumdar, Sayani
Tan, Hongwei
Keywords
ferroelectric tunnel junction, memristor, neuromorphic computing, resistive switching, artificial synapse
Other note
Citation