Modelling of Frequency Dependence in Self-Rectifying Analog BiFeO₃ Memristors

Memristors, recognized as the fourth fundamental circuit element, exhibit highly nonlinear dynamic behavior, analog computing capabilities, and significant potential for chaotic systems and complex dynamics. Self-rectifying analog BiFeO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf> (BFO) memristors, with their high on/off ratio, excellent retention, and remarkable endurance, are particularly promising for these applications. In this work, we present the first demonstration of frequency-dependent measurements on BFO memristors. Our results reveal that the on/off ratio of hysteresis in I-V characteristics and the maximum current in the positive bias range significantly decrease with increasing frequency. We formulated a simplified mathematical model of BFO behavior, based on an exponential ionic drift model for the positive bias region and a diffusion/diode-like I-V behavior for the negative bias region, ensuring high simulation speed while achieving substantial agreement with experimental data. By adjusting model parameters to account for higher amplitudes, we achieved improved alignment with experimental data. Future work will focus on refining the model to better accommodate these dynamic processes and further exploring the relationship between applied bias magnitude and device performance, enhancing the predictive dynamical behavior of BFO memristive devices in practical applications.