An analytical model was proposed to simulate the resistive switching (RS) process of electrochemical metallization memory (ECM) device with a dual-layer structure, in which the AgInSbTe (AIST) buffer layer was introduced to combine with the amorphous carbon (a-C) dielectric layer. This dual-layer ECM device presented good memory performance, such as high switching uniformity, however, a suitable model was absent to understand its RS mechanism. The analytical model of Ag/AIST/a-C/Pt device was developed based on the formation and dissolution of the Ag-ion-based CF. Comparing to the common RS model considering one-step set/reset process, the multi-step of set/reset was utilized to fit with the dual-layer structure. The set/reset always occurs in the AIST layer at first step and in the a-C layer subsequently due to the lower active energy and thermal conductivity of AIST layer. Random disturbances were also introduced to simulate the device variation during the cycling process. Eventually, this model can well fit with the experimental data, providing an analytical tool for describing the RS process of ECM device.
Analytical modeling of electrochemical metallization memory device with dual-layer structure of Ag/AgInSbTe/amorphous C/Pt
Ielmini, Daniele;
2020-01-01
Abstract
An analytical model was proposed to simulate the resistive switching (RS) process of electrochemical metallization memory (ECM) device with a dual-layer structure, in which the AgInSbTe (AIST) buffer layer was introduced to combine with the amorphous carbon (a-C) dielectric layer. This dual-layer ECM device presented good memory performance, such as high switching uniformity, however, a suitable model was absent to understand its RS mechanism. The analytical model of Ag/AIST/a-C/Pt device was developed based on the formation and dissolution of the Ag-ion-based CF. Comparing to the common RS model considering one-step set/reset process, the multi-step of set/reset was utilized to fit with the dual-layer structure. The set/reset always occurs in the AIST layer at first step and in the a-C layer subsequently due to the lower active energy and thermal conductivity of AIST layer. Random disturbances were also introduced to simulate the device variation during the cycling process. Eventually, this model can well fit with the experimental data, providing an analytical tool for describing the RS process of ECM device.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
