Desenvolvimento e estudo de memristores baseados em redes de nanofios de SnO2

• Ano de defesa: 2024
• Autor(a) principal: Moisés, Lucas Augusto
• Orientador(a): Chiquito, Adenilson José
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Física - PPGF
• Palavras-chave em Português: Memristor; Rede de nanofios; Dióxido de estanho; Aprendizagem sináptica; Memórias eletrônicas
• Palavras-chave em Inglês: Nanowire network; Tin dioxide; Synaptic learning; Electronic memories
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::FISICA
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/19632

In this work, memristors were produced based on tin dioxide nanowire networks (2). These nanowires were synthesized by the VLS (Vapor-Liquid-Solid) method. The nanowire structural analysis via X-ray diffraction (XRD) showed a single phase of tin dioxide in the rutile phase with tetragonal symmetry and spatial group P42/mm. Scanning electron microscopy (SEM) images showed varying profiles of sizes and thicknesses (from nano to micrometric) of the obtained wires, a variation arising from the self- assembly process in the synthesis. Three different types of memristors devices have been developed; the first and simplest one is based on a nanowire network obtained directly from synthesis, without any kind of wire manipulation, just adding silver electrodes on the nanostructures. This device showed memristive behavior that strongly depends on the atmosphere in which it was located and data from a detailed study of that dependence indicated the resistive switching was linked to the variation of two types of potentials existing in the network. The first one occurs at the contact region between the nanowires that make up the network, while the second comes from the Schottky-type metal- semiconductor junctions existing at the contact region between the nanowires and electrodes. The variation of these potentials occurs due to charge configuration changes at these contact regions, changes arising from carriers’ flow guided by an applied electric field. A variation of the first device was made by adding a polymeric layer over the nanowire network. This device variation presented a more prominent memristive effect and that improvement was attributed to the molecules trapping on the nanostructure surfaces by the polymer, generating a charge state freezing at surface region. The retention data acquired from this memristor showed behavior congruent with the Ebbinghau forgetting curve, indicating potential of this device in synaptic learning application. Emulations of the learning and forgetting cycles were carried out on this device and that potential was confirmed. The second type of device produced, based on just one wire (microwire), was made in order to study anomalous behaviors observed in some first type devices (specifically those without the polymeric layer). This anomalous behavior was reproduced in this second device, which proved to be a memristor, and was attributed to the protons conduction guided by water molecules adsorbed on surface oxygen vacancies. Unlike the first device, the single-wire device presented constant resistive states, presenting two resistive states retention for more than 18 Ks and an ON/OFF ratio of 4.6 between them. The third type device developed in this work consists of a nanowire network deposited via drop-casting on a substrate. The preferred position of the nanowires in that network obtained was perpendicular to the substrate, unlike the wire network present in the first device memristor, and this position favoured the nanostructures interaction with the environment, thus increasing the water molecules adsorption level. This greater water adsorption led this third type to show an electrical behaviour identical to the single-wire memristor device. Like the single-wire device, the third type device showed constant resistance states over time. Data obtained from retention measurements of two resistive states, with an ON/OFF ratio of 27.8 between them, showed its conservation for more than 1400 s. The retention data from the second and third device types made they candidates for applications in electronic memories.