Experimental evaluation of the dynamic route map in the reset transition of memristive ReRAMs
In this paper, we analyze the reset transition in bipolar TiN/Ti/HfO2 (10 nm)/Al2O3(2 nm)/W ReRAM devices using a tool that allows studying the temporal behaviour of these devices. This tool, the Dynamic Route Map (DRM), provides information about the temporal evolution of the state variable that go...
| Autores: | , , , , , , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión aceptada para publicación |
| Fecha de publicación: | 2020 |
| País: | España |
| Institución: | Consejo Superior de Investigaciones Científicas (CSIC) |
| Repositorio: | DIGITAL.CSIC. Repositorio Institucional del CSIC |
| OAI Identifier: | oai:digital.csic.es:10261/378662 |
| Acceso en línea: | http://hdl.handle.net/10261/378662 https://api.elsevier.com/content/abstract/scopus_id/85090973049 |
| Access Level: | acceso abierto |
| Palabra clave: | Dynamic evolution | Experimental measurement | Memristor | Modelling | Phase space http://metadata.un.org/sdg/9 Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation |
| Sumario: | In this paper, we analyze the reset transition in bipolar TiN/Ti/HfO2 (10 nm)/Al2O3(2 nm)/W ReRAM devices using a tool that allows studying the temporal behaviour of these devices. This tool, the Dynamic Route Map (DRM), provides information about the temporal evolution of the state variable that governs the behaviour of the device, thus allowing an increased insight into resistive switching processes. Here, we show that this DRM is a powerful tool, that may help explaining some non intuitive behaviours of memristors, like the difference in the reset voltage when the inputs are from different frequency or shape. Using this tool, this fact can be explained as a different trajectory on a unique surface defining the device. As a first step, we have used two different models, one based on a physical description, and another one based on the mathematical definition of memristor as a non linear relation between charge and flux. We check that similar DRM can be obtained from both models. Additionally, several series of set-reset transitions have been measured using voltage ramps of different slopes. From the measured transitions, the corresponding resistance has been extracted and, assuming conductive filaments (CF) as the switching mechanism, the corresponding CF radius has been calculated. Using these data, we show that explanations from the model are also supported when using experimental data, thus proving the validity of the approach. |
|---|