Kinetic Monte Carlo analysis of data retention in Al:HfO2-based resistive random access memories

Kinetic Monte Carlo resistive random access memory simulations are used to understand different retention experiments performed at several temperatures. The physics behind resistive switching allows to explain experimental results, in particular the degradation of the conductive filaments with temperature. It is observed that competing mechanisms control resistive switching in this type of experiments and the thermal dependencies involved are key to explain the measurements. Besides, the simulation approach allows to analyze the existence of percolation paths in the device dielectric and the conductive filament density and compactness. Finally, the key physical mechanisms are detected and some clues related to the retention performance and possible technology improvements are unveiled.