Back to All Events

BIT’s 3rd Annual World Congress of Smart Materials-2017


bits_3rd_annual_world_congress_of_smart_materials_2017.jpg

http://www.bitcongress.com/wcsm2017/

In the BIT’s 3rd Annual World Congress of Smart Materials-2017, recently held in Bangkok (Thailand), Prof. Suñé, coordinator of the UAB neuromorphics research group gave a keynote speech on the applications of memristors for non-volatile memory and neuromorphic applications, and an invited talk on volume resistive switching in perovskites.

 

Keynote speech

Resistive switching devices for RRAM and neuromorphic applications

Dr. Jordi Suñé*, Enrique Miranda, Mario Lanza, Javier Martín, J.C González-Rosillo, A. Palau, Teresa Puig, Xavier Obradors

chart_1.jpg

Abstract

Memristors are revolutionizing electronics and computing by allowing energy-efficient non-volatile memories, reconfigurable electronics and neuromorphic systems. In this paper, we review the main properties of resistive switching (RS) devices based on both binary Transition Metal Oxides (TMO), where switching takes place in a conducting filament (CF) of atomic dimensions, and Complex Perovskite Oxides (CPO) which take advantage of collective effects such as those underlying the Metal Insulator Transition (MIT) for interface/volume RS. We discuss the physics and performance of different RS materials and use a recently developed compact model based on the logistic hysteron to capture the DC loops found under voltage-ramp application for set and reset. The model is general and versatile and allows dealing with both abrupt and smooth transitions in all the considered materials. Analysis of the RS loops allows highlighting second order effects related to the temperature which is found to mimic the role of Ca2+ in biological synaptic connections. The requirements for the use of memristors for neuromorphic and RRAM applications are discussed and compared.

*Presenting author.

Invited contribution

Volume Switching in Perovskites for RRAM and Neuromorphic Applications

Dr. Jordi Suñé*, E. Miranda, R. Ortega-Hernández, J.C. González-Rosillo, J. Jareño-Cerrulla, M. Lanza, A. Palau, T. Puig, X. Obradors

chart_2.jpg

Abstract

Resistive Random Access Memory (RRAM) is one of the most promising candidates to replace charge-based non-volatile memories because of several reasons: energy barrier required for information retention is intrinsic to the memory material (non-charge-based), significantly improved energy efficiency, outstanding scalability, simple CMOS-compatible fabrication in the BEOL and switching times down to few nanoseconds. Actually, memory materials are expected to have a major role in future electronics and not only for memory storage, but also for data sensing, processing and displaying. In this regard, the term “memtronics” has recently been coined to highlight the ubiquitous presence of memory devices in near –future applications. RRAM are based on the Resistive Switching (RS) effect, where a change in the resistance of the material can be reversibly induced upon the application of an electric field. In this regard, strongly correlated complex oxides present unique intrinsic properties and extreme sensitivity to external perturbations, which make them idoneous for downscaling switching energy. In particular, metallic complex oxides displaying metal-insulator transition (MIT) are very attractive materials for applications and are barely explored as activeRS elements. In this work, we analyze the RS behavior of three different families of metallic perovskites: La0.8Sr0.2MnO3, YBa2Cu3O7-δ and NdNiO3. We demonstrate that these mixed electronic-ionic conductors undergo a MIT upon the application of an electric field, being able to transform the bulk volume significantly changing its electrical resistance. This volume RS is different in nature from interfacial or filamentary type and opens new possibilities of robust device design. As an example, we present a proof-of-principle result from a 3- Terminal configuration with multilevel memory states.

*Presenting author.

Later Event: April 3
Memrisys 2017