Analytical Modeling and Parameter Extraction of High-Performance Low-Power Memristive Device Based on Trilayer Interlinked Graphene Oxide

• Main Authors: Atiyah A.M., Sedghi H., Ahmadi M.T., Rahmani M.
• Other Authors: 59310016900
• Format: Article
• Published: World Scientific 2025
• Subjects: Associative storage; Compact disks; Ferroelectric RAM; Graphene devices; Islanding; Matrix algebra; Membrane technology; Negative temperature coefficient; Partial discharges; Percolation (computer storage); Percolation (solid state); Positive temperature coefficient; Redox reactions; Schottky barrier diodes; Solvent extraction; Surface discharges; System-on-chip; Tetrodes; Graphene nanoribbons; High resistance; High resistance switching; Low resistance; Low resistance switching; Memristor; Resistance switching; Schottky emissions; Trap assisted tunneling; Trilayer graphene nanoribbon; Trilayers; Current voltage characteristics

To overcome the scaling limitations of large-scale charge-storage-based memories, the physical-based analytic model of a memristor with high-performance resistive random-access memory and low power based on Trilayer interlinked graphene oxide (TIGO) with significant charge transport as an active layer is proposed in this research. To this end, the electron transport of the proposed device is investigated with two electrode-limited conduction mechanisms based on Schottky emission (SE) and trap-assisted tunneling (TAT). In the proposed model, electrically driven reduction of oxygen groups makes the formation of sp2 islands across the TIGO layer. The TIGO-like islands operate as intermediate trap sites and help electrons to tunnel from the cathode toward the anode despite being isolated by the disordered sp3-bonded matrix. The existence of vertically aligned trap sites leads to the formation of percolation paths which allows a steady flow of electrons. The conductive path by the redox of TIGO atoms, because of the conversion of sp3 to sp2 oxygen functionalities, is produced, which can be modeled by degenerate region as ON state with the Low resistance switching (LRS). This path is also ruptured by declining the voltage into the reset voltage, which is modeled by the nondegenerate region as OFF state with High resistance switching (HRS). In fact, the resistance state of the proposed memristor can be reversibly switched by modulating the concentration of sp2 islands. To investigate the performance of the device, the density of states, carrier concentration, electrical conductance in the degenerate and nondegenerate regions, current density and current-voltage characteristics are obtained regarding the energy band structure. In order to verify the accuracy of the research, the models of SE and TAT for two bipolar and unipolar switching modes are compared together and a rational agreement is reported in terms of trend and value. Moreover, the effects of equivalent thermal resistance, interlayer distance, temperature and conductive filament evolution on current-voltage characteristic of the device are investigated. In order to determine the accuracy of the proposed analytical method in this study, the proposed SE and TAT models are compared with each other, and an acceptable agreement is observed. Moreover, the physical-based analytic model of the proposed device in comparison with the experimental data of monolayer graphene nanoribbon and trilayer-structured graphene counterparts is investigated for analogous ambient conditions and rational results are observed. The obtained results of the proposed analytical models and figures of merit for the proposed device showed a promising performance of Trilayer graphene nanoribbon (TGN) for high-performance memristor applications. ? 2024 World Scientific Publishing Company.