Recently, the joint optimization of memristive technology and programming algorithms has made it possible to train neural networks using in-memory computing systems. In this context, the new analog filament conductive metal oxide (CMO)/HfOX Redox-based resistive switching memory (ReRAM) is the key technology. Despite the improvements in device performance reported in the literature, the underlying mechanisms of resistive switching are not fully understood. This study presents the first physics-based analytical model for current transport and resistive switching in these devices. Case studies include analog TaOX/HfOX ReRAM devices are considered. Current transport is described by a trap-to-trap tunneling process, followed by resistive switching by modulation of intra-subband defect density in TaO.X It acts as an electric field and temperature limiting layer. The local temperature and electric field distributions are derived from the solution of the electric and heat transfer equations of the 3D finite element ReRAM model. The intermediate resistance state is explained by the gradual modulation of TaO2.X Electrical conductivity changes due to defect density. The drift dynamics of ions during resistance switching are described analytically, allowing the estimation of the defect transfer energy in TaO2.X floor. Additionally, the role of the electrothermal properties of the CMO layer was revealed. The proposed analytical model accurately describes the experimental switching characteristics of analog TaO2.X/HfOX ReRAM devices increase physics understanding and provide the equations needed for circuit simulations incorporating this technology.
Analytical modelling of the transport in analog filamentary conductive-metal-oxide/HfOx ReRAM devices
Related Posts
Add A Comment