Lithium-ion batteries store energy through two electrochemical reactions. In addition to these main reactions, many side reactions are possible. The causes and effects of battery side reactions are usually detrimental, sometimes positive, and almost always very complicated. We investigate here both fundamental and applied aspects of several side reactions using experiments and theory.

In Part I, we develop a new method to characterize the solid-electrolyte-interphase (SEI), a passivating film that is the product of side reactions on the graphite electrode, using ferrocene, a redox shuttle. By comparing ferrocene kinetics in the presence and absence of passivating films, the shuttle functions as an electrochemical probe of the mechanism by which the SEI prevents reaction. We develop our method using a model system of a rotating-disk-electrode (RDE) and glassy carbon. Coupling experimental measurements with a physics-based model aids interpretation of the results. We develop both steady-state and transient methods in the model system before applying the methods to highly oriented pyrolytic graphite (HOPG), which more closely resembles the carbon found in an actual battery. We study the effects of the relative fraction of edge and basal planes, the formation voltage, and the electrolyte anion. We also attempt in-situ atomic-force microscopy to validate our electrochemical measurements.

In Part II, we treat several different overcharge reactions theoretically to explain empirical observations, predict system behavior, and suggest considerations for battery design. We develop a two-dimensional model to explain why lithium deposition occurs preferentially at the edges of electrodes and suggest mitigation techniques. In a separate model, we implement a redox shuttle for overcharge protection. We study how the presence of a redox shuttle affects the macroscale current-potential relations of the battery and optimize the redox potential of the shuttle while considering the tradeoffs between current efficiency and overcharge protection.

Achieving proper water management in proton-exchange-membrane fuel cells (PEMFC) is an ongoing challenge. Current quantitative treatments of two-phase flow in fuel-cell diffusion media (DM) often assume saturation-dependent transport relations. However, these relations are not validated for the complex physical and chemical structure of the DM, and therefore lack predictive capabilities. Characterization of DM and their water-uptake profiles enables a fundamental understanding of the driving forces behind flooding and facilitates targeted improvement in the form of physically representative simulations. The objective of this work is to characterize, analyze, and explain the wettability and water-uptake behavior of fuel-cell components, namely the DM and catalyst layer (CL), based on manufacturing parameters and physiochemical structure, e.g., fiber structure and hydrophobic treatment of the DM, and chemical composition and crack formation of the CL.

A combination of capillary-pressure saturation (PC-S) measurements, visual-imaging, and physical-characterization techniques is used to quantify water-uptake behavior and identify causal factors. On an intuitive level, Pc-S curves show the propensity of a material to uptake or eject water. The PC-S curves demonstrate that DM are neutrally wetting materials that neither spontaneously imbibe nor eject water. DM from various manufacturers exhibit signature features that can be explained partially by visual differences in fiber structure and deposition of a hydrophobic agent, polytetrafluoroethylene (PTFE). Although the initial addition of PTFE improves hydrophobicity of the sample, increasing PTFE loading does not show significant improvements and instead, as evidenced by pore-size-distribution (PSD) measurements, decreases porosity. Systematic studies of the PSDs demonstrate the level of variation that exists within DM and between DM produced by different manufacturers.

To date, there have not been data on the wettability and water-uptake behavior of CLs. Isolated CLs were made in-house and commercially and tested for their PC-S response. CLs have the propensity to be highly hydrophilic and require capillary pressures as low as -80 kPa to eject water. The presence of Pt or surface cracks increases hydrophilicity. These findings suggest that saturation in CLs, especially cracked CLs, may exacerbate poor transport.

Lastly, this work includes early-stage development of a limiting-current measurement that can be used to calculate effective transport properties as a function of saturation. Results indicate that the method is valid, and different DM have higher transport depending on the operating condition. The technique is yet in a formative stage, and this work includes advice and recommendations for operation and design improvements.