Studies of aprotic sodium-oxygen batteries : superoxide degradation kinetics and design of cathode materials

Abstract

High cost and low energy density are the two main factors that limit the development of lithium-ion batteries (LIBs). Sodium oxygen batteries (SOBs) can theoretically overcome these limitations. The replacement of Li with more abundant Na makes it possible to reduce the cost of raw materials effectively. The discharge product of SOBs, NaO2, has a higher energy density than that of LIBs. This thesis is devoted to study the degradation of NaO2 in aprotic SOBs through experiments and modelling, and to design and develop towards high-performance cathode materials that can mitigate NaO2 degradation via nanostructuring. Chapter 1 is an introduction to SOBs and outlines the critical issues of NaO2 degradation. In Chapter 2, the degradation kinetics of NaO2 was studied by a combination of electrochemical techniques and mathematical modelling. NaO2 degradation is one of the core and challenging problems in SOBs, due to the lack of effective in situ and real time characterization techniques. Instead of relying on advanced characterization methods, the unique voltage profile of NaO2 was used to quantitatively determine the discharge time (td) and charge time (tc) during NaO2 formation and depletion through experiments. The mathematical relation of td and tc was derived based on the fundamental assumption that NaO2 degradation follows a first-order reaction, in order for the reaction rate constant and solubility of NaO2 to be determined. This model can ultimately predict Coulombic efficiency (CE) and guide battery operation. Chapter 3 is an extension to the mathematical model in Chapter 2 to include spatial variation of superoxide concentration and investigate diffusion coupled degradation of NaO2. The influence of superoxide diffusion coefficient, NaO2 solidification threshold, and superoxide diffusion distance on CE was thoroughly explored. Various carbon structures have been investigated to relate the nanostructure to degradation of NaO2 with an aim to design a reversible high capacity air-cathode. In Chapter 4, the accelerated degradation of NaO2 in mesoporous carbon cathodes was studied to reveal the effects of cathode structure. The mesopores are centered at 2.7 nm with a narrow distribution. Contrary to common understanding from literature, this thesis shows that mesoporous carbon causes faster degradation of NaO2 than non-porous carbon. Mesopores can limit the particle size of NaO2 during discharge since these are sites within which NaO2 initially deposited. DFT calculations indicate that small NaO2 tends to have a stronger interaction with the solvent, increasing the reaction rate of NaO2. In Chapter 5, covalently bonded graphene foam (GF) was used as cathode in SOBs. The major discharge product was NaO2 with gravimetric discharge capacity of 43227 mAh/g, Coulombic efficiency of 97%, and 271 cycles. The performance is much better than a commercial carbon paper H2315, which is commonly used in literature. A small amount of Na2CO3 was generated on the surface of GF after each cycle, and Na2CO3 is accumulated as the cycle number increases. The complete coverage of electrode surface by Na2CO3 will lead to the failure of cell. DFT calculations show that Na2CO3 can occupy active sites such as -COC- and –C=O functional groups on the surface of GF, increasing overpotentials. The origin of good performance of GF is likely related with its high specific surface area and unique structure.