Defesa Pública da Tese de Doutorado do Mestre em Engenharia Elétrica Thales Exenberger Becker
Data: 15/03/2022 – terça-feira
Horário: 13h30min
A defesa será realizada através da plataforma Mconf com acesso pelo link: https://mconf.ufrgs.br/webconf/00047686
Banca Examinadora:
Prof. Dr. Marcelo Antonio Pavanello – Relator – Departamento de Engenharia Elétrica – Centro Universitário FEI
Profa. Dra. Fernanda Gusmão de Lima Kastensmidt – PPGC – UFRGS
Prof. Dr. Ivan Muller – PPGEE – UFRGS
Pesidente da banca/Orientador: Prof. Dr. Gilson Inácio Wirth – PPGEE – UFRGS
Título da tese: “ELECTRICAL CHARACTERIZATION AND MODELING OF RANDOM TELEGRAPH NOISE IN MIM-LIKE RESISTIVE SWITCHING DEVICES”
Resumo:
“Metal-insulator-metal (MIM-like) resistive switching (RS) devices have been increasingly studied for several modern and traditional applications, such as information storage, stochastic computing, and bio-inspired computing. In this work, a novel model for anomalous Random Telegraph Noise (aRTN) is presented, accounting for the existence of coupling effect between multiple traps regarding current amplitude deviation. We have determined that the contribution of one defect to the current deviation leading to RTN is dependent on the state (i.e., occupied or vacant) of other defects, indicating the presence of coupling effects. A model is proposed to the describe the behavior at low reading voltages (∼ 0.1 V) for both low-resistance state (LRS) and high-resistance state (HRS). The model can be applied to help understanding the dynamics of filament distribution and trapping/de-trapping activity. Additionally, a novel observation of trap acitivity is presented, which results in giant random conductance fluctuations, up to 3 orders of mangnitude, resembling RTN in RS devices based on TiO2, HfO2 and hexagonal boron nitride (h-BN) under reading voltages. Considering this behavior, presented for three different switching materials, we show that this is a quite general phenomenon and that this significant on/off ratio, in reading conditions, is reproducible and beneficial to ensure recognition of device’s two-state in applications such as stochastic computing integrated circuits (ICs). These events were reproducible for all the aforementioned RS
device types in sequential measurements and under different bias conditions.
Keywords: Resistive Switching, Reliability, Random Telegraph Noise, Trap Coupling Model, Giant RTN.”
