Impedance-based state estimation of lithium-ion batteries

This thesis can be divided into two interconnecting sections. The first section is  centered on the development of an internal temperature (IT) and state of health (SoH) estimator for lithium-ion batteries (LIB) through the use of online electrochemical  impedance spectroscopy (EIS) measurements. Correlation analysis on the relationship  between sensitive impedance variables and battery state parameters was also performed  and the strength of this relationship quantified. The second section focuses on the electrochemical performance of the nanostructured Li-alloying materials silicon (Si) and  germanium (Ge) nanowires (NW) for next-generation LIBs. EIS is used throughout in  order to gain insights into the electrochemical properties of the novel materials. Chapters  are arranged as research articles, each with an introductory summary.  

Chapter 3 describes the development of a novel LIB IT estimator using EIS. The  proposed model used a single frequency impedance point, which displayed high  sensitivity to changes in temperature while simultaneously showing low dependence to  the change in state of charge (SoC) and SoH. The model was able to accurately estimate  the internal temperature of commercial LIBs, over the temperature range 10 °C to 55 °C,  achieving an average RMSE of 1.41 °C across 9 data sets. Chapter 4 describes the  development of a SoH estimation model for LIBs. Similar to the previous chapter, the  impedance at a single frequency is used to relate the fade in capacity to the increase in  impedance as the battery aged. The model successfully estimated the capacity of 6  commercial batteries over 300 cycles, with an RMSE of 0.0064 Ah achieved between  actual and estimated capacities. In chapter 5, sensitivity correlation analysis is performed  to identify and quantify the dependence of various equivalent circuit elements and  frequencies to the vital battery parameters of SoC, SoH and IT. Each parameter  demonstrated high sensitivity to particular frequency ranges and parts of impedance, thus  validating the intrinsic relationship each parameter has with impedance. Chapter 6 details  the electrochemical behaviour of graphite (Gr), Si NW and Ge NW electrodes over a  wide range of parameters including temperature, ageing and SoC. It was found that the nanostructured nature of Si NWs and Ge NWs dramatically improved their low  temperature performance compared to micro-sized Gr. The fracturing and regrowth of the  SEI layer during lithiation/delithiation was monitored using the charge transfer resistance  acquired via EIS. Much larger variations in charge transfer resistance were observed as a function of SoC for Si NW and Ge NW electrodes, which was a result of the cracking and  reformation of the solid electrolyte interface (SEI) layer caused by large volume changes during lithiation and delithiation. In chapter 7, the effect of cell configuration on the  performance of Si NW, Si NW:Gr composite and Gr electrodes was explored. The effect  of separator paper, electrolyte, additives and electrochemical test procedures were  assessed. It was shown that the variation in cell design had a major effect on the rate and  cycle performance of each material, demonstrating the key role of each component on the  overall cell performance.