Modeling and simulation of graphene-oxide-based RRAM

Abstract

We propose a conduction model for resistive random-access memory (RRAM) based on graphene oxide (GO). We associate the electron transport mechanism with a multiphonon trap-assisted tunneling (MTAT) model. Pristine GO is electrically insulating due to the presence of sp3-hybridized oxygen functional groups, e.g., hydroxyl, epoxide, carbonyl, and carboxyl groups. Electrically driven reduction of these oxygen groups triggers formation of nanoscale sp2 islands across the oxide layers. These graphene-like islands act as intermediate trap sites and assist electrons to tunnel from the cathode toward the anode despite being isolated by the disordered sp3-bonded matrix. The presence of vertically aligned trap sites leads to the formation of percolation paths that allow a steady flow of electrons. The resistance state of the RRAM device can then be reversibly switched by electrically modulating the concentration of sp2 islands. This model shows good agreement with experimental data; therefore, we regard MTAT as an admissible explanation for the conduction mechanism in GO-based RRAM.