Phase Separating Electrode Materials Chemical Inductors

Klemen Zelič, Igor Mele, Arghya Bhowmik, Tomaž Katrašnik

Abstract

We discover the presence of chemical inductive effects in phase-separating ion intercalation energy storage materials, specifically in lithium iron phosphate (LFP) and lithium titanate oxide (LTO). These materials feature fast (de)intercalation and slow diffusion relaxation phenomena, which are prerequisites for observing such inductive effects. The presented finding is supported by a mechanistic model and analytical reasoning indicating that all equilibrium states lying inside the miscibility gap of the phase-separating material exhibit a strong inductive response in the low-frequency part of the spectrum. We also explain why such inductive effects are not observed outside the miscibility gap. This letter presents the first mechanistic reasoning of previously reported electrode-level experimental observations of inductance during impedance measurements at low currents.

Introduction

Electrochemical Impedance Spectroscopy (EIS) is considered the "gold standard" for probing the internal dynamics of batteries. However, researchers occasionally observe a "low-frequency inductive loop" (a hook in the Nyquist plot) at very low frequencies (mHz range). In most electrochemical systems, these loops are dismissed as experimental noise, such as cable inductance, corrosion, or poorly fitting cell hardware.

Lithium Iron Phosphate (LFP), a widely used cathode material, consistently displays this inductive behavior. For years, the field lacked a rigorous physical model to explain why a battery material would behave like an inductor. This research addresses a fundamental question: Is this inductance a measurement error, or is it an intrinsic "chemical" property of materials that undergo phase separation?

Methodology: Phase-Field Modeling

The researchers employed a Phase-Field Modeling approach to simulate the behavior of a single LFP particle. The model utilizes the Cahn-Hilliard Equation to simulate "uphill" diffusion and the formation of phase boundaries (spinodal decomposition) within the particle, coupled with the Butler-Volmer Equation to model the fast (de)intercalation of lithium ions at the surface.

Phase-field model of LFP impedance and particle states at various lithiation levels.

The team solved these coupled equations on a 2D domain using a finite volume method. EIS was simulated by applying a small periodic sinusoidal flux and calculating the resulting potential response. Beyond simulations, they linearized the governing equations to prove mathematically that the interaction between a fast destabilizing process (intercalation) and a slow stabilizing process (diffusion) satisfies the formal requirements for a "chemical inductor."

Key Insights: The Role of the Miscibility Gap

The study reveals that the inductive effect only appears when the material is at a lithiation level inside its miscibility gap, where it naturally splits into Li-rich and Li-poor phases. The "chemical inductor" arises from the coupling of two processes with vastly different timescales:

  1. Fast (Milliseconds): The (de)intercalation of ions at the particle surface.
  2. Slow (Seconds): The diffusion of ions across the particle to stabilize phase boundaries.

When the material is outside the miscibility gap (nearly empty or nearly full), it behaves as a homogeneous solid solution. In this state, no slow phase-boundary movement occurs, and the inductive loop disappears, leaving only standard capacitive behavior.

Key Results

Simulations clearly demonstrate inductive loops (negative imaginary impedance) at low frequencies for lithiation levels between 0.2 and 0.8. At levels 0.05 and 0.95, which lie outside the miscibility gap, the loops vanish.

Nyquist plots showing low-frequency inductive loops for states within the miscibility gap.

Bode plot analysis confirms that phase-separated states exhibit a distinct inductive phase shift compared to non-phase-separated states. Furthermore, the authors demonstrated that this is not unique to LFP; Lithium Titanate (LTO) shows the same behavior, suggesting it is a universal trait of phase-separating intercalation materials. Materials like NMC or NCA, which lack domain-based phase separation, do not exhibit this effect.

Conclusion and Future Work

This work provides the first mechanistic answer to questions regarding the impedance of phase-separating materials first posed in 2006. While the single-particle model clarifies the fundamental physics, real-world electrodes are complex porous networks where other factors might overlap with the chemical inductive signal. Future work will focus on experimental setups designed to isolate this chemical inductance from hardware-related artifacts and extending the model to 3D domains to explore potential geometric effects.

Summary updated

Abstract

We discover presence of chemical inductive effects in phase separating ion intercalation energy storage materials, specifically in lithium iron phosphate (LFP) and also lithium titanate oxide (LTO). These materials features fast (de)intercalation and slow diffusion relaxation phenomena which are prerequisites for observing such inductive effects. Presented finding is supported by the mechanistic model and analytical reasoning indicating that all equilibrium states that lay inside the miscibility gap of the phase separating material exhibit strong inductive response in the low frequency part of spectrum. We also explain why such inductive effects are not observed outside the miscibility gap. This letter presents the first mechanistic reasoning of previously reported electrode level experimental observation of inductance during impedance measurements at low currents.
BibTeX
@inproceedings{zelic2023phase,
  author = {Klemen Zelič and Igor Mele and Arghya Bhowmik and Tomaž Katrašnik},
  title = {{Phase Separating Electrode Materials Chemical Inductors}},
  booktitle = {Energy},
  year = {2023},
  doi = {10.1016/j.ensm.2023.01.008},
}
Figure