Quantitative Measure of Hysteresis for Memristors Through Explicit Dynamics

Panayiotis S. Georgiou, Sophia N. Yaliraki, Emmanuel M. Drakakis, Mauricio Barahona

Abstract: We introduce a mathematical framework for the analysis of the input-output dynamics of externally driven memristors. We show that, under general assumptions, their dynamics comply with a Bernoulli differential equation and hence can be nonlinearly transformed into a formally solvable linear equation. The Bernoulli formalism, which applies to both charge- and flux-controlled memristors when either current- or voltage-driven, can, in some cases, lead to expressions of the output of the device as an explicit function of the input. We apply our framework to obtain analytical solutions of the i-v characteristics of the recently proposed model of the Hewlett-Packard memristor under three different drives without the need for numerical simulations. Our explicit solutions allow us to identify a dimensionless lumped parameter that combines device-specific parameters with properties of the input drive. This parameter governs the memristive behavior of the device and, consequently, the amount of hysteresis in the i-v. We proceed further by defining formally a quantitative measure for the hysteresis of the device for which we obtain explicit formulas in terms of the aforementioned parameter and we discuss the applicability of the analysis for the design and analysis of memristor devices.

Introduction

Since the physical realization of the memristor by HP Labs in 2008, researchers have struggled to characterize these devices beyond simple "pinched hysteresis" curves. Because a memristor’s resistance depends on the history of the current or voltage applied to it, its behavior is inherently nonlinear and time-dependent. Until recently, most analysis relied on computationally expensive numerical simulations—essentially "sweeping" parameters to see what happens. There was no general mathematical framework to provide analytical (closed-form) solutions, making it difficult for engineers to design circuits with specific, predictable memristive properties or to quantify exactly how much "hysteresis" a device actually possesses.

The Bernoulli Framework for Memristor Dynamics

The researchers identified that the dynamics of a wide class of ideal memristors (whether charge-controlled or flux-controlled) comply with the Bernoulli differential equation. The "magic" of the Bernoulli equation is that it can be nonlinearly transformed into a standard linear differential equation, which is easily solvable.

Governing differential equations and general solutions for various memristor types, highlighting the Bernoulli Differential Equation (BDE) mapping.

As shown in the table above, this mapping applies across different control and drive modes. By identifying this mathematical structure, the authors derived exact formulas for the output current/voltage of the HP memristor model under three standard engineering signals: sinusoidal, bipolar square, and triangular waves.

Analytical Solutions and the Lumped Parameter (β)

Using this framework, the authors identified a dimensionless constant, the lumped parameter ($\beta$), which combines material properties (like ion mobility and film thickness) with drive properties (like frequency and amplitude). This parameter is a powerful tool for designers; instead of tuning multiple physical variables, a researcher only needs to know $\beta$ to understand the device's nonlinearity and memory.

Analytical i-v characteristics and hysteresis loops for the HP memristor model under sinusoidal (a, d, g), trapezoidal (b, e, h), and triangular (c, f, i) drives.

The analytical solutions provide exact trigonometric and algebraic functions for the $i-v$ characteristics, eliminating the need for SPICE simulations for basic characterization. The results show that memristive behavior is enhanced by small device dimensions, high carrier mobility, and low-frequency/high-amplitude drives—all captured within the $\beta$ parameter.

Quantifying Hysteresis as Energy Dissipation

By defining hysteresis through the lens of thermodynamics (work done), the authors move the concept from a visual "shape" on a graph to a measurable physical quantity ($H$). They proposed a formal measure based on the difference in work done as the device traverses the upper and lower branches of its $i-v$ loop.

A key finding was the "50% Rule": for the HP model, the maximum possible hysteresis is roughly 50% of the energy dissipated by an equivalent linear resistor. This suggests a fundamental limit on the "memristiveness" of simple linear-drift devices. Under a sinusoidal drive, the maximum normalized hysteresis was calculated to be approximately $0.4972$.

Limitations and Future Work

The authors acknowledge that their analysis is based on the first simplified HP model, which assumes linear ionic drift. Real-world devices often exhibit non-linear dopant kinetics and "windowing" effects where boundary movement slows near the device edges. Future work will focus on extending the Bernoulli framework to these more complex models, applying analytical solutions to networks of memristors to observe emergent collective behavior, and accounting for circuit parasitics in high-speed applications.

Summary updated

Abstract

We introduce a mathematical framework for the analysis of the input-output dynamics of externally driven memristors. We show that, under general assumptions, their dynamics comply with a Bernoulli differential equation and hence can be nonlinearly transformed into a formally solvable linear equation. The Bernoulli formalism, which applies to both charge- and flux-controlled memristors when either current- or voltage-driven, can, in some cases, lead to expressions of the output of the device as an explicit function of the input. We apply our framework to obtain analytical solutions of the i-v characteristics of the recently proposed model of the Hewlett-Packard memristor under three different drives without the need for numerical simulations. Our explicit solutions allow us to identify a dimensionless lumped parameter that combines device-specific parameters with properties of the input drive. This parameter governs the memristive behavior of the device and, consequently, the amount of hysteresis in the i-v. We proceed further by defining formally a quantitative measure for the hysteresis of the device for which we obtain explicit formulas in terms of the aforementioned parameter and we discuss the applicability of the analysis for the design and analysis of memristor devices.
BibTeX
@inproceedings{georgiou2012quantitative,
  author = {Panayiotis S. Georgiou and Sophia N. Yaliraki and Emmanuel M. Drakakis and Mauricio Barahona},
  title = {{Quantitative Measure of Hysteresis for Memristors Through Explicit Dynamics}},
  booktitle = {Proceedings of the Royal Society A Mathematical Physical and Engineering Sciences},
  year = {2012},
  doi = {10.1098/rspa.2011.0585},
}

Loading related content...

Figure