A Simple Model for an Intrinsic Piezoelectric Memristor

Piezoelectric memristors represent a convergent frontier in advanced materials science, merging the mechanical-to-electrical transduction properties of piezoelectric materials with the nonlinear, history-dependent resistance behavior characteristic of memristive devices. By combining the inherent ability of piezoelectric materials to convert mechanical deformation into electrical signals with the programmable, nonvolatile resistance switching of memristors, these hybrid components transcend the limitations of conventional single-function devices, offering an integrated platform for concurrent sensing, actuation, and information storage. This convergence is particularly consequential for the field of neuromorphic engineering, where replicating the dynamic plasticity of biological synapses requires components that can detect, respond to, and durably encode incoming signals—a set of demands that piezoelectric memristors are uniquely positioned to fulfill within a single material stack. Beyond cognitive computing architectures, these devices introduce transformative possibilities for energy-autonomous systems, leveraging piezoelectricity to scavenge kinetic energy from environmental sources, such as structural vibrations, human motion, or pressure fluctuations, thereby powering resistive switching operations in the complete absence of conventional energy supplies. In this work, a minimal theoretical framework for an intrinsic piezoelectric memristor is introduced, grounded in the physics of a dimerized one-dimensional chain. The model is reduced to an effective Rice–Mele Hamiltonian, a standard viewpoint for ferroelectric research, which provides both analytical tractability and physical transparency. The numerical simulations are consistent with those reported for polarization-switching features and introduce new dynamical timescales. The stability of the memristive response is also of great significance for neuromorphic and reservoir computing applications.