Wave digital model of a TiN/Ti/HfO<sub>2</sub>/TiN memristor

Enver Solan, Eduardo Pérez, Dennis Michaelis, Christian Wenger, Karlheinz Ochs

International Journal of Numerical Modelling Electronic Networks Devices and Fields · 2019 · 12 citations · 27 references

Concepts

Abstract

Abstract Memristors—nonlinear resistors with memory—are potential candidates for the utilization in self‐organizing circuits. These novel elements need innovative technological developments in order to incorporate them into integrated electrical circuits. Alternatively, HfO 2 ‐based resistive random access memories (RRAMs) are complementary metal‐oxide‐semiconductor (CMOS) compatible and can also be interpreted as memristive devices. A fingerprint of such devices is their rapid change between the high‐ and low‐resistance state. It is intended to exploit this feature in self‐organizing circuits to achieve a desired overall functionality. But the huge device variabilities exacerbate preinvestigations of real circuits including such devices. To this end, a wave digital model based on a mathematical and a physical model of a hafnium oxide (HfO 2 ) memristor is introduced. Utilizing the wave digital approach yields a flexible, real‐time capable algorithmic model, which is suitable for live parameter fitting in real circuits. Comparisons of the emulated hysteresis with measured data of RRAM cells verify the functionality of the presented emulator. The proposed method for emulating physical systems is not restricted to single devices. Indeed, whole subcircuits can be emulated before fabrication in order to save development costs.

References

27