Charge capacity characteristics of a Lithium Nickel-Cobalt-Aluminium Oxide battery show fractional-derivative behavior

Marcus T. Wilson, Vance Farrow, Caleb Pyne, Jonathan ScottarXiv 2021 · arXiv

Introduction

Standard battery models often rely on integer-order differential equations, treating components like resistors and capacitors as ideal elements. However, modern Lithium Nickel-Cobalt-Aluminium Oxide (NCA) batteries exhibit complex "memory" effects where capacity is not a fixed value but a function of discharge history and rate. This phenomenon, often described empirically by Peukert’s Law, suggests that batteries behave as fractional-order systems. This paper explores the physical basis of these effects using ultra-low frequency impedance spectroscopy and fractional calculus.

Methodology

The study utilized a 4.8 Ah NCA battery, a chemistry prevalent in electric vehicles. To capture the long-term memory effects, the researchers performed Ultra-Low Frequency Impedance Spectroscopy (ULFIS) at frequencies as low as 0.5 µHz. A single measurement cycle at this frequency takes approximately 23 days, providing a unique window into the battery's deep-time dynamics.

Impedance spectroscopy data showing magnitude, phase, and Nyquist plots.

The battery was modeled using a Constant Phase Element (CPE) in series with a resistor ($R_s$). Unlike a standard capacitor, a CPE represents a fractional-order integrator where the current-voltage relationship is defined by a fractional derivative of order $\alpha$ (where $0 < \alpha < 1$). To simulate this behavior in standard software, a Morrison RC network—a parallel array of 61 RC branches with logarithmically distributed time constants—was employed.

Key Insights: The Fractional Nature of Peukert’s Law

A significant contribution of this work is the mathematical bridge between fractional calculus and Peukert’s Law. The researchers demonstrated that Peukert’s empirical exponent ($n$) is the reciprocal of the fractional derivative order ($\alpha$): $n = 1/\alpha$.

Experimental results validated this connection. The fractional order $\alpha$ derived from the 23-day impedance plots was 0.976, while the $\alpha$ calculated from capacity-vs-rate discharge tests was 0.971. This high degree of consistency across independent experimental methods confirms that the "capacity offset" observed in batteries is a direct manifestation of fractional-derivative behavior.

Results and Discussion

The NCA battery's fractional order of approximately 0.97 indicates it is very close to, but distinct from, an ideal capacitor ($\alpha = 1$). This slight deviation accounts for the significant "up-tick" in capacity observed at extremely low discharge currents.

Experimental vs. simulated voltage traces at 1 A and 0.1 A.

While the Morrison RC network accurately replicated the general discharge trends, some discrepancies remained. At low currents (e.g., 0.1 A), the experimental data showed small voltage fluctuations or "wiggles." These are likely due to the microscopic structural changes as lithium ions intercalate into the carbon electrodes—effects not captured by the current linear fractional model. Additionally, the model struggled to predict the rapid voltage drop-off as the battery reached full discharge, highlighting the inherent non-linearity of chemical cells.

Conclusion

By treating the NCA battery as a fractional-order system, researchers have provided a physical explanation for century-old empirical observations. This approach allows for more accurate long-term capacity predictions and highlights the importance of "memory" in battery management systems. Future work will focus on integrating non-linear elements and diffusion-based Warburg components to further refine the model's fidelity across diverse battery chemistries.

Summary updated

Abstract

Batteries experience capacity offset where available charge depends on the rate at which this charge is drawn. In this work we analyze the capacity offset of a 4.8 A h lithium nickel-cobalt-aluminium oxide battery using an equivalent circuit model of a fractional capacitor in series with a resistor. In this case, the available charge, in theory, becomes infinite in the limit of infinitesimal rate. We show that the fractional properties of the capacitor can be extracted from the charge against rate plot. We then use a network of RC elements to represent the fractional capacitor in order to simulate the data with Matlab. We find that the fractional exponent alpha obtained in this way, 0.971, agrees with that obtained in a more traditional manner from an impedance versus frequency plot, although the fractional capacity does not. Such an approach demonstrates the importance of a fractional description for capacity offset even when an element is nearly a pure capacitor and is valuable for predictions of state-of-charge when low currents are drawn.
BibTeX
@misc{wilson2021charge,
  author = {Marcus T. Wilson and Vance Farrow and Caleb Pyne and Jonathan Scott},
  title = {{Charge capacity characteristics of a Lithium Nickel-Cobalt-Aluminium Oxide battery show fractional-derivative behavior}},
  year = {2021},
  eprint = {2110.03883},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2110.03883},
}
Figure